GTSAM 4.3.0

[dellaert]

GTSAM 4.3 is a substantial release focused on faster inference, improved correctness and robustness, expanded navigation and estimation capabilities, certifiable and GPU-accelerated optimization, and a more modern developer experience.

Release 4.3 also comes with a substantially expanded MyST documentation site, including new user guides, conceptual documentation, and a growing collection of executable Python notebooks. The examples now cover geometry, SLAM, structure from motion, navigation, filtering, discrete and hybrid inference, continuous-time estimation, certifiable estimation, CUDA optimization, and more, with many notebooks runnable directly in Google Colab.

Contributors

GTSAM 4.3 reflects an unusually broad community effort. The release was driven especially strongly by @dellaert, @varunagrawal, and @ProfFan, with major sustained contributions from @talregev, @p-zach, @Gold856, @DLuminary, @jlblancoc, @jashshah999, and @akshay-krishnan. Their work spans core algorithms, correctness and performance, navigation, geometry, wrappers, documentation and examples, build and CI infrastructure, portability, and modernization of the codebase. The sections below also credit the contributors responsible for major new capabilities.

Major changes in 4.3

Modern C++ and reduced Boost dependence

GTSAM 4.3 moves to C++17 and modern Eigen versions while removing a large amount of legacy Boost usage.

Smart pointers, optionals, tuples, iterators, containers, concepts infrastructure, headers, and other internals were modernized, and code deprecated in GTSAM 4.2 was removed. Boost can now be disabled in substantially more configurations, and support for system-installed dependencies has improved considerably.

Key contributors: @dellaert, @kartikarcot, @varunagrawal, @Gold856, @jlblancoc, @mcm001, @ShuangLiu1992, @jmackay2, @ProfFan.

Faster inference and optimization

A significant amount of work in 4.3 went into making common GTSAM workloads faster. Nonlinear factor-graph error evaluation, Jacobian operations, Levenberg–Marquardt elimination, Lie-group operations, discrete inference, iSAM2 updates, IMU integration, smart factors, and several other hot paths were optimized.

A new multifrontal solver infrastructure adds scheduled and parallel elimination, bottom-up clique merging, task schedulers, parallel separator updates, TBB-based execution, and more memory-efficient parent updates. Additional work introduced matrix-free PCG, compact Schur-complement machinery for bundle adjustment, optional CHOLMOD support, and continuous benchmarking to catch performance regressions.

Key contributors: @dellaert, @ProfFan, @tzvist, @leolrg, @jashshah999, @varunagrawal.

Experimental CUDA acceleration

GTSAM 4.3 introduces experimental CUDA acceleration for both bundle adjustment and more general nonlinear optimization.

The CUDA SFM optimizer provides GPU-backed Levenberg–Marquardt for BAL-style bundle adjustment, CUDA GNC uses the GPU solver as its inner optimizer, and newer CUDA nonlinear solvers support both cuDSS and PCG backends. Python bindings are available for the CUDA LM and GNC paths.

Key contributors: @leolrg, @dellaert.

Navigation, IMU, and filtering expanded substantially

Navigation is one of the areas with the largest amount of new functionality. Highlights include exact zero-order-hold IMU integration, NavState-based Lie-group preintegration, Galilean preintegration and combined Galilean IMU factors, improved handling of sensor offsets and centripetal acceleration, and exact rotating-Earth dynamics.

The release also adds richer NavState Lie-group operations, Galilean-state formulations, EKF and invariant-EKF infrastructure, equivariant filtering and EqVIO support, improved AHRS handling, EKF reset support, gravity-aware IMU factors, and several legged-state-estimation implementations. IMU covariance conventions and residual coordinates also received substantial correctness work.

Key contributors: @dellaert, @scottiyio, @jenniferoum, @rohan-bansal, @mkielo3, @nkhedekar, @arihantb2, @DLuminary, @varunagrawal.

GNSS support became a first-class capability

GTSAM 4.3 significantly expands GNSS estimation support. New factors cover pseudorange, differential pseudorange, carrier phase, RTK double differences, PPP-style measurements, Doppler/range-rate measurements, lever arms, and GNSS/IMU coupling.

The release also adds a GlobalPositioner abstraction and correct Sagnac handling for undifferenced GNSS measurements, with follow-up fixes to the Doppler formulation.

Key contributors: @inuex35, @masoug, @kathirgounder, @mnissov, @scottiyio, @varunagrawal.

Continuous-time Gaussian processes and differentiable splines

A new continuous-time Gaussian-process framework supports white-noise-on-acceleration trajectory models, continuous-time factors, GP interpolation, trajectory factor-graph construction, Python bindings, and worked notebook examples.

GTSAM 4.3 also adds Lie-group-aware differentiable spline curves, extending continuous-time trajectory modeling beyond GP interpolation.

Key contributors: @holmesco, @dellaert.

Constrained and certifiable optimization

Constrained optimization expanded substantially, with support for LP, QP, and QCQP formulations, wrappers, examples, and improved covariance handling under constraints.

QCQP factor graphs can be lifted into optimization problems with monolithic and chordal SDP relaxations, Burer–Monteiro formulations, and recovery back to geometric states. The new certifiable module includes a Riemannian Staircase solver, MOSEK integration in Python and MATLAB, and certifiable applications including Wahba, landmark localization, and rotation-related estimation problems.

FAST-Sync initialization and Shonan certificate handling were also improved.

Key contributors: @dellaert, @yetongumich, @zhexin1904, @avinashresearch1, @ProfFan.

Robust estimation

Robust estimation gained new loss functions, broader GNC support, TLS losses including the René Vidal formulation, GNC-enabled trajectory alignment, and robust CUDA optimization.

The release also includes riSAM, a robust incremental smoothing and mapping implementation built on extensions to iSAM2.

Key contributors: @DanMcGann, @hkhanuja, @akshay-krishnan, @leolrg, @dellaert, @ProfFan.

Better incremental inference, marginalization, and covariance recovery

iSAM2 and the fixed-lag smoothers received numerous fixes and improvements, particularly around marginalization, factor removal, reintroduced variables, adaptive reordering, constrained gradients, tree statistics, timestamps, and state cleanup.

IncrementalFixedLagSmoother moved from gtsam_unstable into the stable library. Marginals and covariance recovery were substantially reworked as well, including Bayes-tree cache handling, Steiner-tree covariance queries, direct covariance recovery through GaussianBayesTree and iSAM2, preservation of joint-marginal key order, and fixes for deep-tree recursion.

Key contributors: @dellaert, @varunagrawal, @gradyrw, @ProfFan, @NewThinker-Jeffrey, @jashshah999, @Ellon, @DLuminary, @tandede, @talfeiner-mc, @admin123-abc.

Lie groups and geometry received a major overhaul

The SO(3), SE(3), and SE₂(3) exponential/logarithm machinery and Jacobians were reworked around faster and more systematic kernels. SO(3) logarithms received both accuracy and performance improvements, especially near difficult rotations.

New or substantially expanded abstractions include Galilean3, SL4, PowerLieGroup, variable-dimension product groups, generic tangent groups, extended SE_k(3) poses, and improved Similarity2/Similarity3. Geometry additions also include Sim3 trajectory alignment, spherical-camera support, fundamental and essential transfer factors, self-calibration factors, and additional QCQP lifts for geometric states.

Key contributors: @dellaert, @mkielo3, @yluo5820, @akshay-krishnan, @AlessandroFornasier, @DLuminary, @inuex35, @kathirgounder.

Hybrid and discrete inference matured considerably

The hybrid and discrete inference stack received a broad overhaul covering factors, conditionals, Bayes nets, nonlinear inference, pruning, sampling, marginalization, smoothing, relinearization, incremental inference, and Python wrapping.

TableFactor provides an efficient sparse representation for discrete factors, while DecisionTree operations, discrete elimination, k-best search, discrete-continuous smoothing and mapping, and hybrid iSAM were substantially improved. Discrete constraints were promoted to the stable API, with new sparse AllDiff support and faster generic multiplication through TableFactor.

Key contributors: @varunagrawal, @dellaert, @ProfFan, @ywkim0606, @arutkowski.

Python, MATLAB, wrappers, packaging, and build support

The Python interface received extensive attention, including broader API coverage, generated type hints, PEP 561 support, NumPy compatibility updates, newer pybind11 integration, better return policies, reduced copies across the wrapper boundary, and many new examples.

A particularly large late-release wrapper push filled gaps in navigation, iSAM2, fixed-lag smoothing, noise-model introspection, subset extraction, CustomFactor, and stub generation.

MATLAB gains CustomFactor, numerical derivatives, revamped bindings, and MOSEK SDP support. Wheel generation, Windows support, Apple Silicon builds, vcpkg integration, CMake configuration, ROS/colcon behavior, and CI reliability were also substantially improved.

Key contributors: @ProfFan, @DLuminary, @varunagrawal, @p-zach, @talregev, @Gold856, @yambati03, @mvanhorn, @zcjhao, @thatdudegrantt, @dellaert, @jlblancoc.

Important correctness and robustness improvements

This release contains substantial correctness work across core inference, marginalization, fixed-lag smoothing, iSAM2, robust objectives, noise models, geometry, IMU preintegration, Jacobians, constraints, serialization, wrappers, and numerical edge cases.

Noise-model validation was tightened, long-standing geometry and Jacobian issues were fixed, robust objective and linearization behavior was made more consistent, covariance-frame conventions were documented, Lie-group Jacobians in prior and between factors are now enabled correctly by default, and numerous memory-safety, serialization, and platform-specific bugs were addressed.

Key contributors: @dellaert, @ProfFan, @varunagrawal, @DanMcGann, @jashshah999, @DLuminary, @gradyrw, and many others.

Documentation and examples

The new MyST-based documentation substantially expands coverage of inference, SLAM, SFM, navigation, geometry, hybrid and discrete models, constrained and certifiable optimization, CUDA solvers, basis functions, marginals, continuous-time Gaussian processes, and the linear solver stack.

Examples and notebooks were reorganized by GTSAM module, with new walkthrough notebooks for wrapped APIs and dedicated examples for newer functionality such as continuous-time estimation, certifiable optimization, navigation, and CUDA solvers.

Key contributors: @p-zach, @dellaert, @CodeXTL, @Robert-Jia00129, @rohan-bansal, @holmesco, @truher, @zcjhao, and many others.

Notable (merged) PRs:

• Adding failing tests for ISAM2 marginalization by @gradyrw in Adding failing tests for ISAM2 marginalization #1105
• Header discipline in base by @dellaert in Header discipline in base #1108
• Windows Tests by @varunagrawal in Windows Tests #1109
• Add missing header for timing target by @dellaert in Add missing header for timing target #1112
• replace ifndefs with pragma once by @acxz in replace ifndefs with pragma once #1113
• Include guards by @acxz in Include guards #1116
• Allow more functionality through system METIS by @acxz in Allow more functionality through system METIS #1117
• Using a random number generator in translation averaging by @akshay-krishnan in Using a random number generator in translation averaging #1120
• Added more tests for basis by @dellaert in Added more tests for basis #1121
• filename in repo is w10000 not w10000-odom by @acxz in filename in repo is w10000 not w10000-odom #1123
• Python optimizer iteration hook by @gchenfc in Python optimizer iteration hook #1127
• Wrap ParameterMatrix as argument for Values by @varunagrawal in Wrap ParameterMatrix as argument for Values #1136
• Apply DecisionTree op with Assignment by @varunagrawal in Apply DecisionTree op with Assignment #1137
• Fixed custom factors in doc/gtsam.lyx/pdf by @dellaert in Fixed custom factors in doc/gtsam.lyx/pdf #1139
• Print parents when printing GaussianConditional by @varunagrawal in Print parents when printing GaussianConditional #1143
• DecisionTree Improvements by @varunagrawal in DecisionTree Improvements #1144
• Wrapping Updates by @varunagrawal in Wrapping Updates #1146
• Improve Gaussian Conditional printing by @varunagrawal in Improve Gaussian Conditional printing #1147
• Pose3 improvements by @dellaert in Pose3 improvements #1148
• Update Wrapper by @varunagrawal in Update Wrapper #1149
• Pose2::Align by @dellaert in Pose2::Align #1150
• DecisionTreeFactor prune method by @varunagrawal in DecisionTreeFactor prune method #1151
• Record number of assignments for each leaf by @varunagrawal in Record number of assignments for each leaf #1154
• DecisionTree Refactor by @varunagrawal in DecisionTree Refactor #1155
• New maxNrAssignment scheme for pruning by @varunagrawal in New maxNrAssignment scheme for pruning #1156
• added wrapper for ProjectionFactorRollingShutter by @shteren1 in added wrapper for ProjectionFactorRollingShutter #1159
• Add camera and Calibration Jacobians in Python by @ProfFan in Add camera and Calibration Jacobians in Python #1171
• noise models: Add getters for all model parameters by @jlblancoc in noise models: Add getters for all model parameters #1175
• Regenerate python utils files upon edit by @gchenfc in Regenerate python utils files upon edit #1176
• Added some utilities for better using the SfM parts of GTSAM in Python by @ProfFan in Added some utilities for better using the SfM parts of GTSAM in Python #1185
• Adding options to translation averaging to support betweenfactor translation priors by @akshay-krishnan in Adding options to translation averaging to support betweenfactor translation priors #1190
• All uncertainty ellipses now contain 95% by @dellaert in All uncertainty ellipses now contain 95% #1193
• Wrap triangulateSafe by @dellaert in Wrap triangulateSafe #1194
• Remove user flag for pip install by @varunagrawal in Remove user flag for pip install #1195
• Hybrid Factor Graph Implementation by @ProfFan in Hybrid Factor Graph Implementation #1203
• Hybrid Pruning by @varunagrawal in Hybrid Pruning #1215
• Improvements by @varunagrawal in Improvements #1217
• Add various tests for Hybrid by @varunagrawal in Add various tests for Hybrid #1220
• HybridFactorGraph base class and other niceties by @varunagrawal in HybridFactorGraph base class and other niceties #1221
• Address matplotlib deprecation warnings for fig.gca() and window.set_title() by @vik748 in Address matplotlib deprecation warnings for fig.gca() and window.set_title() #1225
• Replace boost::function with std::function by @varunagrawal in Replace boost::function with std::function #1232
• added jacobians for all pose3 methods in python wrappers by @shteren1 in added jacobians for all pose3 methods in python wrappers #1234
• Wrapper Update by @varunagrawal in Wrapper Update #1235
• Jose's updates by @varunagrawal in Jose's updates #1237
• Capitalize Identity trait by @varunagrawal in Capitalize Identity trait #1238
• Expose GNC params to python by @amadoantonini in Expose GNC params to python #1239
• Wrap BearingRange Factors by @varunagrawal in Wrap BearingRange Factors #1240
• Implementation of the LOST triangulation algorithm by @akshay-krishnan in Implementation of the LOST triangulation algorithm #1241
• Clear ConcurrentMap before loading archive data by @ScottMcMichael in Clear ConcurrentMap before loading archive data #1248
• Update macos CI env by @varunagrawal in Update macos CI env #1251
• replace addtogroup with ingroup for the SLAM group by @agilemapper in replace addtogroup with ingroup for the SLAM group #1252
• Break up Python CI steps so reading logs is easier by @varunagrawal in Break up Python CI steps so reading logs is easier #1254
• Doxygen Updates by @varunagrawal in Doxygen Updates #1255
• Link to interface library in case system metis is used by @Tobias-Fischer in Link to interface library in case system metis is used #1258
• Delete FindBoost.cmake by @Tobias-Fischer in Delete FindBoost.cmake #1260
• Nonlinear Hybrid by @varunagrawal in Nonlinear Hybrid #1263
• Add Zenodo for citations by @varunagrawal in Add Zenodo for citations #1265
• Hybrid Incremental Inference by @varunagrawal in Hybrid Incremental Inference #1267
• Update README.md by @pglira in Update README.md #1268
• Linear HybridBayesNet optimization by @xsj01 in Linear HybridBayesNet optimization #1270
• Hybrid Incremental by @varunagrawal in Hybrid Incremental #1271
• Hybrid ISAM by @varunagrawal in Hybrid ISAM #1273
• Upgrade from ubuntu 18.04 to 20.04 by @varunagrawal in Upgrade from ubuntu 18.04 to 20.04 #1276
• Additional Nonlinear Hybrid by @varunagrawal in Additional Nonlinear Hybrid #1277
• Hybrid Bayes Net/Tree Optimize by @varunagrawal in Hybrid Bayes Net/Tree Optimize #1280
• Add wrapping for hybrid nonlinear by @ProfFan in Add wrapping for hybrid nonlinear #1281
• Hybrid Bayes Tree Optimize by @varunagrawal in Hybrid Bayes Tree Optimize #1282
• Various updates for Hybrid by @varunagrawal in Various updates for Hybrid #1284
• Rewrap KeyVector for Matlab wrapper by @varunagrawal in Rewrap KeyVector for Matlab wrapper #1285
• Hybrid Serialization by @varunagrawal in Hybrid Serialization #1286
• MixtureFactor continuous keys check by @varunagrawal in MixtureFactor continuous keys check #1289
• Hybrid Improvements by @varunagrawal in Hybrid Improvements #1290
• Improved BayesTree pruning by @varunagrawal in Improved BayesTree pruning #1293
• Hybrid Improvements - II by @varunagrawal in Hybrid Improvements - II #1294
• Check for native architecture and set GTSAM_COMPILE_OPTIONS_PUBLIC accordingly by @stefangachter in Check for native architecture and set GTSAM_COMPILE_OPTIONS_PUBLIC accordingly #1296
• Hybrid Bayes Net pruning by @varunagrawal in Hybrid Bayes Net pruning #1300
• Hybrid Gaussian Conditional Elimination by @varunagrawal in Hybrid Gaussian Conditional Elimination #1301
• Update geometry.i : wrap multiplication of Rot3 and Unit3 by @akshay-krishnan in Update geometry.i : wrap multiplication of Rot3 and Unit3  #1302
• Use eigen3 config.cmake by @OznOg in Use eigen3 config.cmake #1304
• Hybrid Updates by @varunagrawal in Hybrid Updates #1306
• Hybrid Pruning by @varunagrawal in Hybrid Pruning #1309
• Update Switching fixture to be 0 indexed by @varunagrawal in Update Switching fixture to be 0 indexed #1311
• Dsf-gtsfm-refactor by @dellaert in Dsf-gtsfm-refactor #1312
• Hybrid ISAM GetOrdering method by @varunagrawal in Hybrid ISAM GetOrdering method #1315
• Wrapper Update by @varunagrawal in Wrapper Update #1317
• Hybrid Mixture Error by @varunagrawal in Hybrid Mixture Error #1318
• Proper Hybrid Elimination by @varunagrawal in Proper Hybrid Elimination #1319
• Housekeeping Updates by @varunagrawal in Housekeeping Updates #1321
• Hybrid Multifrontal by @varunagrawal in Hybrid Multifrontal #1323
• Remove unnecessary functions in ImuFactorsExample.cpp by @Taeyoung96 in Remove unnecessary functions in ImuFactorsExample.cpp #1324
• add workflow to trigger ubuntu packaging by @berndpfrommer in add workflow to trigger ubuntu packaging #1328
• Add an expression for normalize() by @nkhedekar in Add an expression for normalize() #1330
• Update README.md by @MigVega in Update README.md #1331
• Hybrid Elimination by @varunagrawal in Hybrid Elimination #1339
• Fixed Jacobian in python CustomFactorExample by @ulterzlw in Fixed Jacobian in python CustomFactorExample #1340
• Hybrid Improvements by @varunagrawal in Hybrid Improvements #1343
• Replace NoiseModelFactor1 in with NoiseModelFactorN in pre-made factors by @gchenfc in Replace NoiseModelFactor1 in with NoiseModelFactorN in pre-made factors #1344
• GaussianConditional logDeterminant by @varunagrawal in GaussianConditional logDeterminant #1345
• Sampling test for Hybrid Posterior by @varunagrawal in Sampling test for Hybrid Posterior #1346
• Hybrid Bayes Net sampling by @varunagrawal in Hybrid Bayes Net sampling #1347
• Some refactoring and new test by @dellaert in Some refactoring and new test #1349
• Add logDensity and evaluate to GaussianBN and HybridBN by @dellaert in Add logDensity and evaluate to GaussianBN and HybridBN #1352
• Added convenience constructors and python wrappers by @dellaert in Added convenience constructors and python wrappers #1353
• Wrap constructors for PreintegratedAhrsMeasurements by @varunagrawal in Wrap constructors for PreintegratedAhrsMeasurements #1354
• HybridValues specify continuous then discrete by @varunagrawal in HybridValues specify continuous then discrete #1355
• Likelihoods and other improvements by @dellaert in Likelihoods and other improvements #1356
• Test on ratio by @dellaert in Test on ratio #1357
• Initializer list and insert/update chaining by @dellaert in Initializer list and insert/update chaining #1359
• Hybrid Elimination with Normalization Constant by @varunagrawal in Hybrid Elimination with Normalization Constant #1360
• Slightly expand Python API by @jrpowers in Slightly expand Python API #1361
• Switch to EliminateDiscrete for max-product by @dellaert in Switch to EliminateDiscrete for max-product #1362
• Fixed a few small issues... by @dellaert in Fixed a few small issues... #1367
• Hybrid Serialization by @varunagrawal in Hybrid Serialization #1368
• Un-deprecate NoiseModelFactor1-6 and move X1 and key1-6 shortcuts to NoiseModelFactorN by @gchenfc in Un-deprecate NoiseModelFactor1-6 and move X1 and key1-6 shortcuts to NoiseModelFactorN #1370
• Two new tests, one with 5 modes by @dellaert in Two new tests, one with 5 modes #1371
• Simplified AP for HybridBayesNet by @dellaert in Simplified AP for HybridBayesNet #1372
• DefaultOrderingFunc in EliminationTraits by @varunagrawal in DefaultOrderingFunc in EliminationTraits #1373
• Big Re-"Factor" of Hybrid by @dellaert in Big Re-"Factor" of Hybrid #1374
• Get rid of boost::assign by @dellaert in Get rid of boost::assign #1375
• More Expression wrappers by @BrettRD in More Expression wrappers #1376
• replace list_of with initializer list in TimeTBB.cpp by @kartikarcot in replace list_of with initializer list in TimeTBB.cpp #1379
• Harmonized error/logProbability in entire Factor/Conditional hierarchy by @dellaert in Harmonized error/logProbability in entire Factor/Conditional hierarchy #1380
• BatchFixedLagSmoother: add ability to limit warning debug prints by @acowley in BatchFixedLagSmoother: add ability to limit warning debug prints #1381
• Some Hybrid Improvements by @varunagrawal in Some Hybrid Improvements #1387
• Hybrid/simplify by @dellaert in Hybrid/simplify #1388
• Correction to docs and lambdas by @varunagrawal in Correction to docs and lambdas #1390
• Switch pruning test to probabilities by @varunagrawal in Switch pruning test to probabilities #1391
• verdant/typedef-optional-references by @kartikarcot in verdant/typedef-optional-references #1392
• Switch pruning test to probabilities. by @dellaert in Switch pruning test to probabilities. #1394
• Switch to version 4.3a0 and to c++17. by @dellaert in Switch to version 4.3a0 and to c++17. #1395
• Deprecate filter by @dellaert in Deprecate filter #1397
• Deprecate filter in 4.3 by @dellaert in Deprecate filter in 4.3 #1398
• Remove code deprecated in 4.2 by @dellaert in Remove code deprecated in 4.2 #1399
• Verdant/replace boost optional vals by @kartikarcot in Verdant/replace boost optional vals #1403
• Upgraded to Eigen 3.4.0 by @dellaert in Upgraded to Eigen 3.4.0 #1405
• Verdant/smart ptrs by @kartikarcot in Verdant/smart ptrs #1406
• Deprecated boost transform_iterators by @dellaert in Deprecated boost transform_iterators #1407
• Removed deprecated boost transform_iterators by @dellaert in Removed deprecated boost transform_iterators #1408
• Verdant/boost serialization by @kartikarcot in Verdant/boost serialization #1409
• Verdant/boost tuples by @kartikarcot in Verdant/boost tuples #1410
• Replace boost::ptr_map in Values... by @dellaert in Replace boost::ptr_map in Values... #1411
• Iterator for 4.2 by @dellaert in Iterator for 4.2 #1417
• Added a const_iterator back in by @dellaert in Added a const_iterator back in #1418
• Update ROS package.xml format to v3 and update gtsam version (develop branch) by @jlblancoc in Update ROS package.xml format to v3 and update gtsam version (develop branch) #1428
• Add maintainers to package xml by @jlblancoc in Add maintainers to package xml #1430
• More details on source licenses by @jlblancoc in More details on source licenses #1432
• c++17 style pairs everywhere by @dellaert in c++17 style pairs everywhere #1437
• Getting rid of misc. boost usage by @dellaert in Getting rid of misc. boost usage #1438
• replace boost phoenix parser with SignatureParser by @dellaert in replace boost phoenix parser with SignatureParser #1439
• Remove 'boost' dependency from Lago by @ankur-rc in Remove 'boost' dependency from Lago #1440
• Fixing stack overflow issue for large tree by @wanmeihuali in Fixing stack overflow issue for large tree #1441
• Refactor elimination... by @dellaert in Refactor elimination... #1442
• Separate .i file for Values by @varunagrawal in Separate .i file for Values #1443
• Optionally remove boost completely by @dellaert in Optionally remove boost completely #1444
• Update CppUnitLite to not rely on boost by @varunagrawal in Update CppUnitLite to not rely on boost #1447
• Add normalization to max-product, avoiding underflow. by @keevindoherty in Add normalization to max-product, avoiding underflow. #1449
• Add normalization to sum-product, avoiding underflow. by @keevindoherty in Add normalization to sum-product, avoiding underflow. #1453
• Fixed warnings that arise from stricter compiler flags by @dellaert in Fixed warnings that arise from stricter compiler flags #1456
• Cross-platform traceStorage by @dellaert in Cross-platform traceStorage #1458
• Reverted ill-advised concepts, attempt 2 by @dellaert in Reverted ill-advised concepts, attempt 2 #1460
• Allow override of BUILD_SHARED_LIBS by @Ibarria in Allow override of BUILD_SHARED_LIBS #1464
• CMakes updates by @varunagrawal in CMakes updates #1465
• Better Hybrid support by @varunagrawal in Better Hybrid support #1466
• Cmake Compatibility by @varunagrawal in Cmake Compatibility #1469
• Wrapper Update for GTSAM 4.3 by @varunagrawal in Wrapper Update for GTSAM 4.3 #1474
• Stable BatchFixedLagSmoother by @dellaert in Stable BatchFixedLagSmoother #1475
• Add constant term for hybrid models post-hoc by @varunagrawal in Add constant term for hybrid models post-hoc #1481
• CI Updates by @varunagrawal in CI Updates #1484
• Update CodingGuidelines.lyx by @jlblancoc in Update CodingGuidelines.lyx #1485
• cmake: explicitly specifying the linker language for gtsam by @jlblancoc in cmake: explicitly specifying the linker language for gtsam #1486
• Allow using system pybind11 if it exists. by @jlblancoc in Allow using system pybind11 if it exists. #1487
• GTSAM_DT_MERGING Flag by @varunagrawal in GTSAM_DT_MERGING Flag #1501
• Some refine works related to range-based for-loop of map constainer by @zubingtan in Some refine works related to range-based for-loop of map constainer #1502
• Add Pose2 Component Jacobians by @DanMcGann in Add Pose2 Component Jacobians #1504
• add clang-format configuration file by @zubingtan in add clang-format configuration file #1506
• add useLOST to triangulateSafe by @dllu in add useLOST to triangulateSafe #1517
• Add missing option in the CMake for Pose2 Expmap by @ProfFan in Add missing option in the CMake for Pose2 Expmap #1520
• Upgrade TBB by @varunagrawal in Upgrade TBB #1523
• Wrap Unit3 methods with Jacobians by @varunagrawal in Wrap Unit3 methods with Jacobians #1524
• Added TableFactor, a discrete factor optimized for sparsity. by @ywkim0606 in Added TableFactor, a discrete factor optimized for sparsity. #1528
• Wapper Update by @varunagrawal in Wapper Update #1530
• LOST Triangulation Underconstrained by @travisdriver in LOST Triangulation Underconstrained #1534
• CMake Updates by @varunagrawal in CMake Updates #1536
• Basis Module Improvements by @varunagrawal in Basis Module Improvements #1537
• Decision Tree Improvements by @varunagrawal in Decision Tree Improvements #1542
• Pybind11 STL Support by @varunagrawal in Pybind11 STL Support #1546
• Make Basis factors dynamic by @varunagrawal in Make Basis factors dynamic #1548
• Python Wrapper Backwards Compatibility by @varunagrawal in Python Wrapper Backwards Compatibility #1549
• Remove boost from TableFactor and added guards to testSerializationSlam by @arutkowski in Remove boost from TableFactor and added guards to testSerializationSlam #1552
• Add explanation of the StereoPoint2 constructor by @IshitaTakeshi in Add explanation of the StereoPoint2 constructor #1554
• Remove need for nrAssignments in DecisionTreeFactor::prune by @varunagrawal in Remove need for nrAssignments in DecisionTreeFactor::prune #1555
• TableFactor Improvements by @varunagrawal in TableFactor Improvements #1556
• Overload +=, , operator by @varunagrawal in Overload +=, , operator #1558
• Add non concurrency to all workflows by @talregev in Add non concurrency to all workflows #1564
• Handle latest tbb by @talregev in Handle latest tbb #1567
• Wrap ISAM2 full update method by @varunagrawal in Wrap ISAM2 full update method #1569
• Enable GNC Optimizer by @varunagrawal in Enable GNC Optimizer #1572
• Improved documentation, wrapper, and tests by @dellaert in Improved documentation, wrapper, and tests #1574
• Hybrid Elimination Improvements by @varunagrawal in Hybrid Elimination Improvements #1575
• Added evaluation of expressions when inserting in Values by @arutkowski in Added evaluation of expressions when inserting in Values #1577
• Show remaining keys in InconsistentEliminationRequested by @varunagrawal in Show remaining keys in InconsistentEliminationRequested #1578
• Improved wrapper/tests/documentation for some discrete classes by @dellaert in Improved wrapper/tests/documentation for some discrete classes #1579
• TableFactor Unary Apply Methods by @varunagrawal in TableFactor Unary Apply Methods #1580
• Added some python and c++ examples with BinaryMeasurements by @dellaert in Added some python and c++ examples with BinaryMeasurements #1583
• Additional quaternion test by @dellaert in Additional quaternion test #1589
• Efficient Discrete Elimination by @varunagrawal in Efficient Discrete Elimination #1590
• TableFactor Constructors by @varunagrawal in TableFactor Constructors #1592
• Additional Helpers by @varunagrawal in Additional Helpers #1594
• Add some new loss functions by @NaivelyWritten in Add some new loss functions #1597
• Simplify scripts by @talregev in Simplify scripts #1598
• Add ninja build by @talregev in Add ninja build #1609
• Add Lie group capabilities to NavState by @varunagrawal in Add Lie group capabilities to NavState #1613
• Hybrid Document Update by @varunagrawal in Hybrid Document Update #1617
• Update Python CustomFactor documentation and example by @cntaylor in Update Python CustomFactor documentation and example #1618
• Python Wrapper Updates by @varunagrawal in Python Wrapper Updates #1628
• Remove setting CMAKE_MACOSX_RPATH to 0 by default by @traversaro in Remove setting CMAKE_MACOSX_RPATH to 0 by default #1632
• Add cpp example for GNC by @achintyamohan in Add cpp example for GNC #1634
• Update INSTALL.md by @stepeos in Update INSTALL.md #1646
• Matlab Wrapper Update by @varunagrawal in Matlab Wrapper Update #1649
• Templated FactorGraph::at method by @varunagrawal in Templated FactorGraph::at method #1650
• CMake Cleanup by @varunagrawal in CMake Cleanup #1651
• Global DefaultKeyFormatter Assignment by @varunagrawal in Global DefaultKeyFormatter Assignment #1655
• Overload JacobianFactor::getA method by @varunagrawal in Overload JacobianFactor::getA method #1656
• Re-enable Debug CI for Linux by @varunagrawal in Re-enable Debug CI for Linux #1658
• Chebyshev2 Improvements by @varunagrawal in Chebyshev2 Improvements #1660
• DiscreteFactor::errorTree method for all assignments by @varunagrawal in DiscreteFactor::errorTree method for all assignments #1669
• printErrors method for HybridFactorGraph by @varunagrawal in printErrors method for HybridFactorGraph #1670
• Optimizer Improvements by @varunagrawal in Optimizer Improvements #1671
• Add a missing field in Matlab's CMakeLists.txt for UNSTABLE build by @etamarlu in Add a missing field in Matlab's CMakeLists.txt for UNSTABLE build #1677
• Add reprojection with fisheye camera model by @kkelchte in Add reprojection with fisheye camera model #1678
• Compile gtsam python for windows by @talregev in Compile gtsam python for windows #1685
• AlgebraicDecisionTree Helpers by @varunagrawal in AlgebraicDecisionTree Helpers #1696
• Cleaner version of eliminate by @dellaert in Cleaner version of eliminate #1698
• Cephes Updates by @varunagrawal in Cephes Updates #1704
• Print logNormalizationConstant for Gaussian conditionals by @varunagrawal in Print logNormalizationConstant for Gaussian conditionals #1705
• ISAM2: Add variables before computing delta by @varunagrawal in ISAM2: Add variables before computing delta #1707
• libcephes only needs 3.10, allowing build on ubuntu 18.04 by @berndpfrommer in libcephes only needs 3.10, allowing build on ubuntu 18.04 #1711
• Update Wrapper by @varunagrawal in Update Wrapper #1712
• Various Improvements by @varunagrawal in Various Improvements #1713
• Remove swap from python ci by @talregev in Remove swap from python ci #1714
• Support msvc 2022 along 2019 by @talregev in Support msvc 2022 along 2019 #1715
• Ubuntu 22.04 for python tbb. hopefully will solve the hang problem. by @talregev in Ubuntu 22.04 for python tbb. hopefully will solve the hang problem. #1719
• Fixing std::optional handling being redundant in Boost version > 1.80 by @talregev in Fixing std::optional handling being redundant in Boost version > 1.80 #1721
• Close unmatched endif by @stevenlovegrove in Close unmatched endif #1728
• Specialize BetweenFactor with Similarity3 for Python bindings by @niwhsa9 in Specialize BetweenFactor with Similarity3 for Python bindings #1729
• Fixing strength of bias constraints in IMUKittiExampleGPS.cpp by @MichalNowicki in Fixing strength of bias constraints in IMUKittiExampleGPS.cpp #1735
• Removed a self include in PreintegrationCombinedParams.h by @nkhedekar in Removed a self include in PreintegrationCombinedParams.h #1737
• Sort VectorValues::values() by key when TBB is ON by @arutkowski in Sort VectorValues::values() by key when TBB is ON #1740
• Some Updates by @varunagrawal in Some Updates #1743
• Add missing header in 3rdparty/cephes by @dch4o in Add missing header in 3rdparty/cephes #1748
• Tiny Updates to Python Wrapper by @contagon in Tiny Updates to Python Wrapper #1755
• Correctly deal with sensor rotation in AHRS by @dellaert in Correctly deal with sensor rotation in AHRS #1762
• Build all tests for windows by @talregev in Build all tests for windows #1764
• Update MSVC toolset version by @varunagrawal in Update MSVC toolset version #1768
• gtsam ros pkg not propagating dependencies for downstream ros pkgs by @matlabbe in gtsam ros pkg not propagating dependencies for downstream ros pkgs #1772
• Update wrap and allow numpy 2.0.0 by @varunagrawal in Update wrap and allow numpy 2.0.0 #1773
• Update namespace for Vector and Matrix by @varunagrawal in Update namespace for Vector and Matrix #1774
• correct Linux CI platforms by @acxz in correct Linux CI platforms #1775
• modernize containers by @acxz in modernize containers #1780
• DiscreteConditional and DiscreteBayesNet improvements by @varunagrawal in DiscreteConditional and DiscreteBayesNet improvements #1781
• Improve DiscreteConditional's argmax and argmaxInPlace by @varunagrawal in Improve DiscreteConditional's argmax and argmaxInPlace #1785
• add definition of constraints by @yetongumich in add definition of constraints #1789
• Cherry-pick Add Pose2 Component Jacobians #1504 by @ProfFan in Cherry-pick #1504 #1798
• Improvements to GaussianConditional and GaussianBayesNet by @varunagrawal in Improvements to GaussianConditional and GaussianBayesNet #1801
• Working Hybrid by @varunagrawal in Working Hybrid #1802
• Direct Hybrid Factor Specification by @varunagrawal in Direct Hybrid Factor Specification #1805
• Python port of LocalizationExample.cpp by @varunagrawal in Python port of LocalizationExample.cpp #1808
• Use bundled pybind11 for Python wrapper by @varunagrawal in Use bundled pybind11 for Python wrapper #1812
• Generate Python type hints during build by @varunagrawal in Generate Python type hints during build #1813
• add example CameraResectioning.py by @truher in add example CameraResectioning.py #1816
• added SelfCalibrationExample.py by @truher in added SelfCalibrationExample.py #1817
• Update cmake by @dellaert in Update cmake #1818
• Check if cmake policy exists by @dellaert in Check if cmake policy exists #1820
• Use pybind11-stubgen for generating stubs by @ProfFan in Use pybind11-stubgen for generating stubs #1821
• Improved HybridBayesNet API by @varunagrawal in Improved HybridBayesNet API #1823
• Remove gcc9+tbb CI for python wrapper by @varunagrawal in Remove gcc9+tbb CI for python wrapper #1825
• Add the BATA translation averaging factor with unit tests by @akshay-krishnan in Add the BATA translation averaging factor with unit tests #1828
• Hybrid tests using importance sampling by @dellaert in Hybrid tests using importance sampling #1829
• Rename Hybrid Elements by @varunagrawal in Rename Hybrid Elements #1830
• HybridNonlinearFactor Relinearization by @varunagrawal in HybridNonlinearFactor Relinearization #1831
• Enum to categorize HybridFactor by @varunagrawal in Enum to categorize HybridFactor #1832
• Update Hybrid Renaming by @varunagrawal in Update Hybrid Renaming #1833
• Remove HybridNonlinearFactor normalization code by @varunagrawal in Remove HybridNonlinearFactor normalization code #1834
• Split HybridNonlinearFactor into .h and .cpp files by @varunagrawal in Split HybridNonlinearFactor into .h and .cpp files #1835
• HybridGaussianConditional inherits from HybridGaussianFactor by @varunagrawal in HybridGaussianConditional inherits from HybridGaussianFactor #1836
• Common errorTree method and its use in HybridGaussianFactorGraph by @varunagrawal in Common errorTree method and its use in HybridGaussianFactorGraph #1837
• numdiff in python by @dellaert in numdiff in python #1838
• Use k or -log(k) for normalization constant by @varunagrawal in Use k or -log(k) for normalization constant #1839
• Document and favor easy constructor by @dellaert in Document and favor easy constructor #1840
• Non-loop versions for checking sigma values by @varunagrawal in Non-loop versions for checking sigma values #1842
• Easy HNF constructors by @dellaert in Easy HNF constructors #1843
• Small runtime improvement for hybrid by @dellaert in Small runtime improvement for hybrid #1844
• Ignore stubgen errors and wrap TriangulationFactor by @varunagrawal in Ignore stubgen errors and wrap TriangulationFactor #1846
• Moved mixture model tests to their own file... by @dellaert in Moved mixture model tests to their own file... #1847
• Simpler HybridGaussianFoo constructors by @dellaert in Simpler HybridGaussianFoo constructors #1848
• Remove postfix for compiled Python library by @varunagrawal in Remove postfix for compiled Python library #1849
• Better HybridNonlinear! by @dellaert in Better HybridNonlinear! #1851
• More refactoring by @dellaert in More refactoring #1853
• More tests, some small bugfixes by @dellaert in More tests, some small bugfixes #1854
• Navigation docs by @dellaert in Navigation docs #1856
• Implement errorTree in DiscreteFactor by @varunagrawal in Implement errorTree in DiscreteFactor #1858
• Implement errorTree for HybridNonlinearFactorGraph by @varunagrawal in Implement errorTree for HybridNonlinearFactorGraph #1859
• Refactored hybrid inference by @dellaert in Refactored hybrid inference #1863
• Updates to No Hiding PR by @varunagrawal in Updates to No Hiding PR #1865
• Improve the DiscreteFactorFromErrors by @varunagrawal in Improve the DiscreteFactorFromErrors #1867
• precompiled_header.h: remove chartTesting.h by @valgur in precompiled_header.h: remove chartTesting.h #1870
• Faster DecisionTree by @dellaert in Faster DecisionTree #1875
• Improve Linear Conjugate Gradient Method by @varunagrawal in Improve Linear Conjugate Gradient Method #1876
• Refactor NonlinearConjugateGradientOptimizer by @varunagrawal in Refactor NonlinearConjugateGradientOptimizer #1877
• Additional Nonlinear CG Direction Methods by @varunagrawal in Additional Nonlinear CG Direction Methods #1878
• Rosenbrock function for testing Nonlinear Conjugate Gradient Method by @varunagrawal in Rosenbrock function for testing Nonlinear Conjugate Gradient Method #1879
• Significant speedup by @dellaert in Significant speedup #1881
• Some Hybrid iSAM cleanup by @dellaert in Some Hybrid iSAM cleanup #1883
• Two fundamental matrix classes by @dellaert in Two fundamental matrix classes #1884
• TransferFactor with numerical derivatives by @dellaert in TransferFactor with numerical derivatives #1885
• EssentialTransferFactor by @dellaert in EssentialTransferFactor #1887
• F wrapper and bugfix by @dellaert in F wrapper and bugfix #1888
• Allow parent project to override options by @spacey-sooty in Allow parent project to override options #1889
• Clean up Switching tests by @varunagrawal in Clean up Switching tests #1892
• Lots of Hybrid Improvements by @varunagrawal in Lots of Hybrid Improvements #1894
• Update MacOS Images for CI by @varunagrawal in Update MacOS Images for CI #1895
• More comparisons by @dellaert in More comparisons #1896
• Add missing #if for clang by @jlblancoc in Add missing #if for clang #1897
• Reduce memory consumption during discrete pruning by @varunagrawal in Reduce memory consumption during discrete pruning #1898
• Bump min required cmake version to 3.9 by @jlblancoc in Bump min required cmake version to 3.9 #1899
• Make operator-> const by @OznOg in Make operator-> const #1900
• wrap BatchFixedLagSmoother.getFactors() by @truher in wrap BatchFixedLagSmoother.getFactors() #1909
• Remove the asserts as they have already been checked in Eigen by @ProfFan in Remove the asserts as they have already been checked in Eigen #1910
• Change the deprecated C++ enum math to constexprs by @ProfFan in Change the deprecated C++ enum math to constexprs #1911
• Updates based on profiling by @varunagrawal in Updates based on profiling #1912
• add error(HybridValues) to NonlinearFactor by @truher in add error(HybridValues) to NonlinearFactor #1913
• Add CMake switch for GPerfTools by @varunagrawal in Add CMake switch for GPerfTools #1917
• Document disabling tests and examples in INSTALL.md by @spacey-sooty in Document disabling tests and examples in INSTALL.md #1918
• Discrete Elimination Refactor by @varunagrawal in Discrete Elimination Refactor #1919
• Ring struct in a separate file by @varunagrawal in Ring struct in a separate file #1922
• Kalman filter update by @dellaert in Kalman filter update #1924
• Make DiscreteFactor::operator() a common base method by @varunagrawal in Make DiscreteFactor::operator() a common base method #1925
• Common Discrete Ops by @varunagrawal in Common Discrete Ops #1926
• Various Discrete Improvements by @varunagrawal in Various Discrete Improvements #1927
• Endow NavState with group operations by @dellaert in Endow NavState with group operations #1930
• Left and Right SO(3) Jacobians by @dellaert in Left and Right SO(3) Jacobians #1932
• Faster TableFactor::toDecisionTreeFactor by @varunagrawal in Faster TableFactor::toDecisionTreeFactor #1933
• projection and SFM for 2d poses by @truher in projection and SFM for 2d poses #1934
• Better SE(3) and SE2(3) Jacobians by @dellaert in Better SE(3) and SE2(3) Jacobians #1938
• Add serialize() to various preintegration classes by @varunagrawal in Add serialize() to various preintegration classes #1939
• [cephes] use GTSAM_LIBRARY_TYPE by @mcm001 in [cephes] use GTSAM_LIBRARY_TYPE #1942
• Various Improvements by @varunagrawal in Various Improvements #1943
• Improve build time when updating Boost flags by @varunagrawal in Improve build time when updating Boost flags #1944
• Bump jinja2 from 3.1.4 to 3.1.5 in /wrap/pybind11/docs by @app/dependabot in Bump jinja2 from 3.1.4 to 3.1.5 in /wrap/pybind11/docs #1945
• Add BUILD_SHARED_LIBS as a configurable option by @Gold856 in Add BUILD_SHARED_LIBS as a configurable option #1946
• Discrete Improvements by @varunagrawal in Discrete Improvements #1947
• Hybrid Timing by @varunagrawal in Hybrid Timing #1948
• TableFactor and TableDistribution by @varunagrawal in TableFactor and TableDistribution #1953
• TableFactor within Hybrid Elimination by @varunagrawal in TableFactor within Hybrid Elimination #1954
• Customize discrete elimination in Hybrid by @varunagrawal in Customize discrete elimination in Hybrid #1955
• Enable more warnings and remove unused variables by @mcm001 in Enable more warnings and remove unused variables #1959
• TableFactor Serialization by @varunagrawal in TableFactor Serialization #1961
• DiscreteFactor multiply method by @varunagrawal in DiscreteFactor multiply method #1963
• Clean up CMake build by @Gold856 in Clean up CMake build #1964
• Platform specific #ifs by @varunagrawal in Platform specific #ifs #1966
• QNX Compatibility by @JaiXJM-BB in QNX Compatibility #1968
• Remove operator^ from Matrix and SubgraphPreconditioner by @calcmogul in Remove operator^ from Matrix and SubgraphPreconditioner #1969
• Add cassert include to Chebyshev2.cpp by @mcm001 in Add cassert include to Chebyshev2.cpp #1971
• Add missing test include for assert_inequal() by @calcmogul in Add missing test include for assert_inequal() #1973
• Adding lever-arm to GPS factors by @mnissov in Adding lever-arm to GPS factors #1975
• Faster SE(3) and SE_2(3) exponential maps by @dellaert in Faster SE(3) and SE_2(3) exponential maps #1979
• add constrained optimizers by @yetongumich in add constrained optimizers #1980
• Reduce smart pointer copies in destructors by @mcm001 in Reduce smart pointer copies in destructors #1983
• Marginalize Discrete Modes by @varunagrawal in Marginalize Discrete Modes #1985
• Improvements to Hybrid Inference by @varunagrawal in Improvements to Hybrid Inference #1986
• City10000 dataset example by @varunagrawal in City10000 dataset example #1989
• Refactor jointBayesNet by @dellaert in Refactor jointBayesNet #1991
• Better HybridSmoother by @varunagrawal in Better HybridSmoother #1996
• Update installation and build documentation for Windows & Python on Windows by @p-zach in Update installation and build documentation for Windows & Python on Windows #2000
• Silence a warning about a memcpy overread for gcc 13 by @jmackay2 in Silence a warning about a memcpy overread for gcc 13 #2002
• move constructors by @dellaert in move constructors #2003
• Search for K-best solutions by @dellaert in Search for K-best solutions #2005
• Search for k-best in factor graph by @dellaert in Search for k-best in factor graph #2006
• Wrapper for DiscreteSearch by @dellaert in Wrapper for DiscreteSearch #2009
• Updates to City10000 Example by @varunagrawal in Updates to City10000 Example #2010
• kill expNormalize in favor of scale by @dellaert in kill expNormalize in favor of scale #2011
• City1000 script by @dellaert in City1000 script  #2014
• Lots of savings! by @dellaert in Lots of savings! #2016
• Re-linearization by @dellaert in Re-linearization #2017
• Python script for hybrid inference on City10000 dataset by @varunagrawal in Python script for hybrid inference on City10000 dataset #2019
• IMU factor with Gravity Estimation by @nkhedekar in IMU factor with Gravity Estimation #2020
• Set up MyST site deploy by @p-zach in Set up MyST site deploy #2021
• Add overviews for existing docs folders on MyST site by @p-zach in Add overviews for existing docs folders on MyST site #2023
• Edit required check so that it doesn't hang when only non-code files modified by @p-zach in Edit required check so that it doesn't hang when only non-code files modified #2024
• Create doc ipynb for Rot2 by @p-zach in Create doc ipynb for Rot2 #2025
• Improvements to HybridCity10000 python script by @varunagrawal in Improvements to HybridCity10000 python script #2027
• Add cibuildwheel workflow by @yambati03 in Add cibuildwheel workflow #2029
• Add Rot3 documentation by @p-zach in Add Rot3 documentation #2030
• Updated elaboratePoint2KalmanFilter.cpp example and easyPoint2KF bugfix by @mkielo3 in Updated elaboratePoint2KalmanFilter.cpp example and easyPoint2KF bugfix #2031
• Adds Jacobians to Lie::interpolate and others by @BrettRD in Adds Jacobians to Lie::interpolate and others #2035
• Bugfix Bearing/Range Pose3toPose3 Jacobians by @DanMcGann in Bugfix Bearing/Range Pose3toPose3 Jacobians #2039
• Update Rot3 doc by @p-zach in Update Rot3 doc #2041
• Deprecate GTSAM_SLOW_BUT_CORRECT_EXPMAP by @varunagrawal in Deprecate GTSAM_SLOW_BUT_CORRECT_EXPMAP #2043
• Move scripts for building wheels to CI directory by @varunagrawal in Move scripts for building wheels to CI directory #2044
• Hat and Vee by @dellaert in Hat and Vee #2045
• Use GTSAM_CONCEPT_USAGE in lieu of BOOST_CONCEPT_USAGE by @varunagrawal in Use GTSAM_CONCEPT_USAGE in lieu of BOOST_CONCEPT_USAGE #2047
• Wrap Update by @varunagrawal in Wrap Update #2049
• More FrobeniusFactors by @dellaert in More FrobeniusFactors #2052
• Implement Equivariant Filter Bias example in Python by @jenniferoum in Implement Equivariant Filter Bias example in Python #2054
• Make docstring generation accessible by @p-zach in Make docstring generation accessible #2055
• Hybrid Updates by @varunagrawal in Hybrid Updates #2056
• Remove fixed values when keys are reintroduced in newFactors by @varunagrawal in Remove fixed values when keys are reintroduced in newFactors #2061
• Nonlinear Hybrid Smoother by @varunagrawal in Nonlinear Hybrid Smoother #2062
• Exposed ExtendedKalmanFilter to Python with examples by @mkielo3 in Exposed ExtendedKalmanFilter to Python with examples #2063
• CI using Docker by @suveergarg in CI using Docker #2065
• Python bindings: Enable use of Similarity2/Similarity3 in nonlinear optimization by @mhalber in Python bindings: Enable use of Similarity2/Similarity3 in nonlinear optimization #2067
• Clean up Ordering + headers by @dellaert in Clean up Ordering + headers #2068
• Faster Chebyshev2 by @dellaert in Faster Chebyshev2 #2069
• Wrap missing Chebyshev2 methods... by @dellaert in Wrap missing Chebyshev2 methods... #2071
• Update wrap 0327 by @p-zach in Update wrap 0327 #2076
• Discrete-Continuous Smoothing and Mapping by @varunagrawal in Discrete-Continuous Smoothing and Mapping #2078
• Update Metis' cmake_minimum_required to 3.5 by @varunagrawal in Update Metis' cmake_minimum_required to 3.5 #2079
• Doc generation with GPT by @p-zach in Doc generation with GPT #2080
• AHRS docs and wrap by @dellaert in AHRS docs and wrap #2082
• Move IncrementalFixedLagSmoother from unstable to stable by @p-zach in Move IncrementalFixedLagSmoother from unstable to stable #2083
• More navigation docs by @dellaert in More navigation docs #2084
• New docs: inference by @p-zach in New docs: inference #2087
• Remove static identity matrices in lago.cpp to avoid memory corruption in downstream tasks by @BITVoyager in Remove static identity matrices in lago.cpp to avoid memory corruption in downstream tasks #2091
• Trigger doxygen generation in cibuildwheel by @yambati03 in Trigger doxygen generation in cibuildwheel #2092
• Add support for building MacOS wheels in cibuildwheel by @yambati03 in Add support for building MacOS wheels in cibuildwheel #2093
• Update wrap by @yambati03 in Update wrap #2094
• Bump minimum Ubuntu runner version to 22.04 by @yambati03 in Bump minimum Ubuntu runner version to 22.04 #2095
• Add range SLAM example by @dellaert in Add range SLAM example #2096
• Update navbar by @p-zach in Update navbar #2097
• Ordering doc update by @p-zach in Ordering doc update #2098
• Revamped DexpFunctor and LogmapDerivative by @dellaert in Revamped DexpFunctor and LogmapDerivative #2099
• Docs minor updates by @p-zach in Docs minor updates #2100
• Add cibuildwheel targets for MacOS Silicon by @yambati03 in Add cibuildwheel targets for MacOS Silicon #2102
• C++ example of an Equivariant filter by @darshan-17 in C++ example of an Equivariant filter  #2103
• Docs for symbolic folder by @dellaert in Docs for symbolic folder #2105
• Towards fixing Doxygen-to-docstring issues by @p-zach in Towards fixing Doxygen-to-docstring issues #2107
• Add wrapper for imuBias in iSAM2 by @ProfFan in Add wrapper for imuBias in iSAM2 #2108
• Add production workflow for cibuildwheel by @yambati03 in Add production workflow for cibuildwheel #2110
• New docs: SLAM by @p-zach in New docs: SLAM #2111
• An EKF Example by @dellaert in An EKF Example #2114
• Invariant EKF by @dellaert in Invariant EKF #2115
• Fixed KarcherMean, LAGO, PlanarProjectionFactor by @dellaert in Fixed KarcherMean, LAGO, PlanarProjectionFactor #2116
• Working rig ! by @dellaert in Working rig ! #2117
• (Update wrap) add functionality to pull docs from parent classes by @p-zach in (Update wrap) add functionality to pull docs from parent classes #2118
• New docs: hybrid by @p-zach in New docs: hybrid #2120
• Adding Galilean3 Lie group by @mkielo3 in Adding Galilean3 Lie group #2122
• Add script to clean up gtsam-develop project releases by @yambati03 in Add script to clean up gtsam-develop project releases #2123
• Smart factors docs by @dellaert in Smart factors docs #2126
• Allow for dynamic M/G in new EKF variants by @dellaert in Allow for dynamic M/G in new EKF variants #2129
• Move testIncrementalFixedLagSmoother to gtsam by @varunagrawal in Move testIncrementalFixedLagSmoother to gtsam #2132
• Wrap DiscreteMarginals by @varunagrawal in Wrap DiscreteMarginals #2134
• Hybrid Improvements by @varunagrawal in Hybrid Improvements #2135
• Discrete Sampling by @varunagrawal in Discrete Sampling #2136
• Update sample methods by @varunagrawal in Update sample methods #2137
• Add Multi RobotKeyFormatter as an external variable (Windows) by @DrWisdome in Add Multi RobotKeyFormatter as an external variable (Windows) #2140
• docs: Add Invariant Kalman Filtering to navigation.md by @scottiyio in docs: Add Invariant Kalman Filtering to navigation.md #2147
• Update navigation.md by @scottiyio in Update navigation.md #2149
• Updated wrapper for HybridValues by @dellaert in Updated wrapper for HybridValues #2150
• Markdown documentation draft PR for InvariantEKF by @scottiyio in Markdown documentation draft PR for InvariantEKF #2151
• Wrap HybridBayesNet and HybridConditional methods by @dellaert in Wrap HybridBayesNet and HybridConditional methods #2152
• Fixed SLAM Example link in Pose2 Doc by @CodeXTL in Fixed SLAM Example link in Pose2 Doc #2155
• Added user guide for geometry/Cal3_S2 by @CodeXTL in Added user guide for geometry/Cal3_S2 #2156
• Add template for IMU backend selection to move away from compiler flag by @p-zach in Add template for IMU backend selection to move away from compiler flag #2157
• Created jupyter notebook for the LQR example. by @zhangzdd in Created jupyter notebook for the LQR example. #2160
• Updated Cal3_S2 User Guide by @CodeXTL in Updated Cal3_S2 User Guide #2163
• Add vcpkg ci by @talregev in Add vcpkg ci #2164
• ROS package.xml: do not depend on boost by @jlblancoc in ROS package.xml: do not depend on boost #2171
• Removed dimension field from Cal3 by @CodeXTL in Removed dimension field from Cal3 #2178
• Updates to Similarity2 and Similarity3 by @dellaert in Updates to Similarity2 and Similarity3 #2180
• Add GTSAM_USE_SYSTEM_PYBIND option and improve cache for vcpkg by @talregev in Add GTSAM_USE_SYSTEM_PYBIND option and improve cache for vcpkg #2182
• Similarity2 and Similarity3 updates by @dellaert in Similarity2 and Similarity3 updates #2183
• Add Gal3 wrapper and... by @dellaert in Add Gal3 wrapper and... #2184
• Ported StereoVOExample by @CodeXTL in Ported StereoVOExample #2186
• MatrixLieGroup implementation and traits by @dellaert in MatrixLieGroup implementation and traits #2187
• Updated Cal3DS2_Base python wrapper by @CodeXTL in Updated Cal3DS2_Base python wrapper #2188
• Add friendly instructions for custom wrapper extensions by @p-zach in Add friendly instructions for custom wrapper extensions #2189
• Bespoke vec methods by @dellaert in Bespoke vec methods #2192
• Test both PIM variants by @dellaert in Test both PIM variants #2193
• Ported StereoVOExample_large and Edits to StereoVOExample by @CodeXTL in Ported StereoVOExample_large and Edits to StereoVOExample #2195
• iLQR example notebook by @zhangzdd in iLQR example notebook #2196
• EquivariantFilter and GroupAction by @jenniferoum in EquivariantFilter and GroupAction #2197
• Remove initial cov from CombinedImuFactor by @dellaert in Remove initial cov from CombinedImuFactor #2199
• Added User Guides for Cal3_S2Stereo and StereoCamera by @CodeXTL in Added User Guides for Cal3_S2Stereo and StereoCamera #2200
• Add DiscreteScenario by @p-zach in Add DiscreteScenario #2201
• Increase maxIterations in Cal3DS2_Base.cpp calibrate function by @steplee in Increase maxIterations in Cal3DS2_Base.cpp calibrate function #2204
• Add reference to wrap docs by @p-zach in Add reference to wrap docs #2206
• Add SL4 to geometry by @dellaert in Add SL4 to geometry #2207
• Concepts documentation and remove Dim/dim by @dellaert in Concepts documentation and remove Dim/dim #2209
• Refactor AHRSFactor by @dellaert in Refactor AHRSFactor #2210
• Add Gal3 and SL4 docs by @dellaert in Add Gal3 and SL4 docs #2214
• refactor Gal3 + Analytical ExpmapDerivative by @dellaert in refactor Gal3 + Analytical ExpmapDerivative #2215
• Exclude windows tests by @talregev in Exclude windows tests #2220
• Remove cxx_flags by @talregev in Remove cxx_flags #2223
• Kernels for SO3 by @dellaert in Kernels for SO3 #2226
• Use Kernels, deprecate old methods by @dellaert in Use Kernels, deprecate old methods #2227
• Port eval to other repo by @p-zach in Port eval to other repo #2228
• CI Improvements for speed by @dellaert in CI Improvements for speed #2229
• GalileanFactor ! by @dellaert in GalileanFactor ! #2231
• Add gtsam unstable tests in vcpkg by @talregev in Add gtsam unstable tests in vcpkg #2234
• FrobeniusBetweenFactorNL, and better noise API by @dellaert in #2236
• NavState + IMU EKF by @dellaert in #2242
• Remove metis and geographiclib ports from vcpkg ci by @talregev in #2244
• Remove outdated brew tap (was required for boost versions) by @ProfFan in #2245
• Add Gal3 wrapping by @dellaert in #2247
• Documented NonlinearOptimizerParams in the NonlinearOptimizer doc by @rafaelspring in #2250
• Gal3 IMU EKF class & header by @scottiyio in #2251
• Remove GeographicLib as a 3rd party vendor by @warrenyun in #2253
• Discrete docs by @dellaert in #2257
• More notebooks in discrete by @dellaert in #2258
• Add some more examples by @dellaert in #2259
• Exact zero-order hold IMU integration by @dellaert in #2261
• Formatting/naming in Nav by @dellaert in #2263
• ExtendedPriorFactor and ConcentratedGaussian by @dellaert in #2266
• temporary remove eigen3 from vcpkg by @talregev in #2267
• Enable and propagate /permissive- on MSVC by @Gold856 in #2268
• Add vcpkg compilation with gtsam in INSTALL.md doc by @talregev in #2269
• Remove python excluded tests on windows vcpkg by @talregev in #2271
• Update wrap to latest with pybind11 3.0.1 by @ProfFan in #2277
• Add tests to ensure centripetal accel is dealt with properly for IMUs offset from body frame by @scottiyio in #2281
• Add GeographicLib to vcpkg by @talregev in #2282
• Cast Eigen::Matrix dimension type to int by @varunagrawal in #2284
• Documentation Updates by @Robert-Jia00129 in #2285
• Small example for bearing-range by @dellaert in #2286
• Use doublecross instead of cross(Cross(a, b)) in NavState by @scottiyio in #2287
• 6 More Python Examples + Standardized Notebook Format by @rohan-bansal in #2288
• Add back eigen3 to vcpkg by @talregev in #2289
• Added discrete_sreries example tutorial by @Robert-Jia00129 in #2290
• Remove temporary eigen3 by @talregev in #2293
• Remove cephes dllexport.h from cmake by @talregev in #2294
• [1/N] Split the NoiseModelFactorN to individual header by @ProfFan in #2301
• [2/N] Change NoiseModelFactorN to be an alias to NoiseModelFactorT by @ProfFan in #2302
• Update version to 4.3a1 by @dellaert in #2303
• Remove-macos-13-ci by @dellaert in #2304
• Remove absolute paths in configuration installs by @ethanmusser in #2308
• add python exports for SmartStereoProjectionFactor by @dvorak0 in #2313
• Various filter changes by @dellaert in #2315
• Added PowerLieGroup by @dellaert in #2316
• Added BayesTree cache clearing for Marginals by @travisdriver in #2318
• Remove try-catch from large tangent vector tests by @app/copilot-swe-agent in #2320
• MultifrontalSolver by @dellaert in #2327
• Address code review comments: documentation and naming conventions by @app/copilot-swe-agent in #2328
• JacobianFactor now compute the error more efficiently by first assembling the X vector by @ProfFan in #2329
• Bottom-up clique merging and TBB by @dellaert in #2330
• Handle noise and NonlinearEquality constraints by @dellaert in #2332
• Allow for HessianFactors in MultifrontalSolver by @dellaert in #2333
• Feature/marginals by @dellaert in #2334
• TBB for solve by @dellaert in #2336
• Cache optimization by @dellaert in #2339
• Optimize Levenberg Marquardt elimination by caching JunctionTree. by @tzvist in #2340
• Jacobian implementations for sim3 transformFrom by @akshay-krishnan in #2341
• Task scheduler and ForestTraversal by @dellaert in #2342
• Scheduled MultifrontalSolver by @dellaert in #2343
• Address code review feedback for PriorityScheduler and ForestTraversal by @app/copilot-swe-agent in #2344
• Two Schedulers - One very fast Solver by @dellaert in #2345
• range/bearing for NavState and Gal3 by @dellaert in #2346
• Parallel separator update in MFSolver by @dellaert in #2347
• Minimize the number of workflows by @varunagrawal in #2349
• Class for Pose3 trajectory alignment using Sim3 by @akshay-krishnan in #2351
• Address code review feedback for TrajectoryAlignerSim3 by @app/copilot-swe-agent in #2352
• Sampler updates by @dellaert in #2353
• perf: parallelize NonlinearFactorGraph::error computation. by @tzvist in #2355
• perf: TBB with memory growth control. by @tzvist in #2356
• Tiny perf improvements by @dellaert in #2358
• Add vcpkg mkl by @talregev in #2359
• switch Hat/Vee to orthonormal sl(4) basis and update adjoint/tests by @yluo5820 in #2360
• deltaError replaces 2 error calls by @dellaert in #2361
• Nonlinear multifrontal solver by @dellaert in #2362
• Default noise models and TranslationRecovery patch by @dellaert in #2363
• Revert linear solver type to MULTIFRONTAL_CHOLESKY by @akshay-krishnan in #2365
• sccache for all os in vcpkg by @talregev in #2367
• Expose EqF Python bindings (ABC), template on N, modify/add examples by @rohan-bansal in #2368
• Simple GNSS Pseudorange Factor by @masoug in #2369
• Run Windows CI workflow on push by @Gold856 in #2370
• Remove all Cephes source files except those needed for igami by @Gold856 in #2371
• EKF Reset by @dellaert in #2373
• Docs for basis by @dellaert in #2375
• Small issues in examples by @dellaert in #2376
• Standardize on KeySet by @dellaert in #2377
• Introducing DifferentialPseudorangeFactor by @masoug in #2383
• Add continuous benchmarking workflow by @ProfFan in #2385
• Use vcpkg manifest by @Gold856 in #2388
• Constrained docs by @dellaert in #2394
• Update SFMBAL benchmark script by @ProfFan in #2397
• NavState docs by @dellaert in #2398
• Clean up Boost dependencies by @Gold856 in #2399
• Clean up and speed up CI by @Gold856 in #2400
• removed unnecessary DiscreteKeys constructor by @aminoh in #2402
• Remove macOS 14 from CI by @Gold856 in #2406
• Optimize cibuildwheel build times by @Gold856 in #2407
• iSAM2 Updates to Support riSAM Implementation by @DanMcGann in #2408
• Robust Incremental Smoothing and Mapping (riSAM) by @DanMcGann in #2409
• Allowing Non Noise Model Factors for Gnc by @hkhanuja in #2410
• Continuous-Time Gaussian Process Factors (Factors and Interpolation) by @holmesco in #2411
• Weight update in Gnc GM through static function in LossFunctions by @hkhanuja in #2415
• Extended Pose SE_k(3) by @dellaert in #2416
• TLS loss and Rene Vidal TLS implementation by @hkhanuja in #2417
• Harmonize Ad/ad by @dellaert in #2418
• Address PR #2418 review feedback: SWIG Jacobians, trait-based dim, AdjointMap perf note by @app/copilot-swe-agent in #2419
• TrajectoryAlignerSim3 - add point3 constraints if available, optionally use GNC by @akshay-krishnan in #2445
• Add SphericalCamera Python bindings by @inuex35 in #2446
• Add PseudorangeFactorArm and DifferentialPseudorangeFactorArm by @inuex35 in #2447
• Add BAL SmartFactor example by @dvorak0 in #2449
• variable dimension product groups by @dellaert in #2450
• Remove benchmark caching by @ProfFan in #2451
• Reduce wheel CI flakiness by @dellaert in #2452
• Wrap DCSAM class and remove hybrid examples by @varunagrawal in #2454
• Add EqVIO Group, Manifold Foundation by @rohan-bansal in #2455
• Use internal MyST links in published docs by @dellaert in #2458
• Hybrid improvements by @varunagrawal in #2459
• Move free functions definitions to GncOptimizer.cpp by @varunagrawal in #2461
• Reject negative sigma values in noise model constructors by @jashshah999 in #2465
• 4 legged state estimators by @dellaert in #2470
• Rename Python reserved keywords in wrapper interface files by @jashshah999 in #2471
• EqVIO Filter Implementation by @rohan-bansal in #2475
• Steiner trees for Covariance Recovery by @dellaert in #2476
• Add optional ecef_T_nav to PseudorangeFactorArm for GNSS/IMU coupling by @inuex35 in #2477
• Simplify Marginals and add docs by @dellaert in #2482
• [codex] Move covariance recovery into GaussianBayesTree and expose ISAM2 queries by @dellaert in #2486
• MATLAB CustomFactor by @dellaert in #2491
• numerical derivatives in MATLAB by @dellaert in #2493
• Add group action param to Product Lie Group by @rohan-bansal in #2494
• Added GlobalPositioner Class by @kathirgounder in #2495
• Add missing templates in wrapper by @dellaert in #2496
• [codex] Refactor RangeFactor to remove expression overhead by @dellaert in #2497
• Simplify SAM factors by @dellaert in #2498
• Generalize AttitudeFactor by @dellaert in #2499
• New CustomFactor examples by @dellaert in #2501
• feat: double difference factor for rtk gnss by @inuex35 in #2502
• Geometry enhancements by @dellaert in #2503
• Add python wrapper and test for LocalOrientedPlane3Factor by @SchmidL in #2507
• feat(isam2): adaptive variable reordering triggered by fill-in growth by @jashshah999 in #2509
• perf(linear): compute JacobianFactor::gradient() directly, skip HessianFactor conversion by @jashshah999 in #2510
• Continuous Time GP Factor Graph by @holmesco in #2514
• navigation: honor configured IMU bias in LeggedFixedLagSmoother by @ChiyunNoh in #2515
• Images from ChatGpt by @dellaert in #2518
• perf(python): return references from optimizer to avoid deep-copying Values by @jashshah999 in #2520
• WNOA CT-GP Bindings and Example by @holmesco in #2523
• Factor for yaw in Pose2, with bias by @truher in #2524
• LP and QP by @dellaert in #2525
• Add QCQP by @dellaert in #2526
• LP/QP/QCQP wrapper and examples by @dellaert in #2527
• Use sccache for Windows Python jobs by @Gold856 in #2530
• Remove unused Boost dependencies by @Gold856 in #2532
• More edits for gcc 13 by @kshaji3 in #2538
• Modify for QCQP support for the BM methods. by @zhexin1904 in #2541
• cmake: disable Boost by default when building inside colcon (ROS 2) by @jlblancoc in #2543
• [Python] Add venv setup instructions for PEP 668 systems in README by @zcjhao in #2546
• DecisionTree::dot(name, ...): skip system() shellout on iOS by @ShuangLiu1992 in #2547
• Add FromQcqpValue (Rot2/Rot3) and ExtractQcqpValues helpers in feature/convert_to_QCQP by @zhexin1904 in #2548
• WIP: add homogenization term to D = 1 lifting to QCQP factor by @avinashresearch1 in #2549
• Add QCQP factor graph conversion by @dellaert in #2550
• Copy-free matrix views for wrapped batch APIs by @dellaert in #2551
• feat(python): Expose BetweenFactor to Python wrapper by @SidArora29 in #2554
• Optimize compile times by @Gold856 in #2561
• feat(python): expose PriorFactor to Python wrapper by @SidArora29 in #2562
• Add SelfCalibrationFactor for closed-form focal-length self-calibration by @kathirgounder in #2568
• Generic TangentGroup TG from action-parameterized ProductLieGroup by @AlessandroFornasier in #2570
• [Python] Add batch triangulateSafe supporting incomplete tracks by @zcjhao in #2573
• navigation: route undifferenced GNSS factors through Sagnac-corrected geodist (#2563) by @inuex35 in #2574
• navigation: add undifferenced (PPP) pseudorange and carrier-phase factors by @inuex35 in #2575
• Add compiler hash to vcpkg cache key by @Gold856 in #2577
• Clean up headers/includes and refresh precompiled headers by @Gold856 in #2578
• Remove aligned operator new macro by @Gold856 in #2579
• Add experimental CUDA SFM bundle adjustment optimizer by @leolrg in #2581
• [codex] Automate PyPI develop cleanup by @dellaert in #2583
• BatchFactor by @dellaert in #2586
• Sync/update wrap subtree by @thatdudegrantt in #2587
• navigation: add GNSS Doppler (range-rate) factors by @inuex35 in #2589
• Link cephes-gtsam against libm by @hz2 in #2590
• Use Matrix initializer_list constructor instead of comma initializers by @Gold856 in #2591
• Clarify JointMarginal::at()/operator() indexing semantics (Key, not position) by @thatdudegrantt in #2592
• CUDA GNC: run GncOptimizer with the CUDA SFM LM inner solver, ~20x over CPU GNC by @leolrg in #2593
• Fixed lag and python by @bboyack in #2599
• Avoid recursive separator marginal caching for deep Bayes trees by @admin123-abc in #2600
• Notebooks by @zcjhao in #2604
• Implements monolithic and chordal SDP from QCQP intermediate by @avinashresearch1 in #2606
• Add pragmas for GCC 16 warnings by @mjelen2 in #2607
• Add certifiable module and Riemannian Staircase solver by @dellaert in #2608
• Lifting Rot3, Pose2 and Pose3 to QCQP (for D = 1) and monolithic/chordal SDP including recovery and examples. by @avinashresearch1 in #2609
• Modernize installation and repository guidance by @dellaert in #2611
• New wrap and MOSEK wrapping by @dellaert in #2612
• Certifiable Wahba Problem and Landmark Localization by @avinashresearch1 in #2616
• Speed up geometry by @dellaert in #2617
• Update version to 4.3a2 by @dellaert in #2618
• Clean up Doxygen generation and speed up Python CI by @dellaert in #2619
• Refine certifiable QCQP recovery, examples, and Python docs by @dellaert in #2623
• Try preventing vcpkg resource exhaustion by @Gold856 in #2624
• PsyCh Example by @dellaert in #2625
• Optimize ProductLieGroup and tangent-group Jacobians by @dellaert in #2627
• Accelerate iterative PCG with parallel matrix-free operators by @ProfFan in #2628
• METIS: Avoid redundant system linking & use public API for separator partitioning by @ilkecan in #2631
• Allow using system-installed Spectra by @ilkecan in #2632
• Speed up PR CI with nightly compatibility matrices by @dellaert in #2633
• Add templated FAST-Sync initializer by @dellaert in #2634
• Preserve joint marginal query key order by @dellaert in #2636
• Run no-Boost C++ tests in required Python CI by @dellaert in #2637
• Refactor timimg by @dellaert in #2638
• Initialize Shonan averaging with FAST-Sync by @dellaert in #2639
• Build examples in nightly C++ CI by @dellaert in #2640
• Refactor smart projection factors around shared base by @dellaert in #2641
• Handle constrained components in ISAM2 gradients by @dellaert in #2643
• Deprecate legacy unstable IMU state APIs by @dellaert in #2644
• Update bundled pybind11 to 3.0.4 by @dellaert in #2647
• Remove orphaned KITTI MATLAB examples by @dellaert in #2648
• Rename indeterminate system exception and add debugging guide by @dellaert in #2649
• Document covariance frames for between and prior factors by @dellaert in #2650
• Automatically linearize fixed-size binary and ternary factors by @dellaert in #2652
• Use normalized all-pivot Cholesky checks by @dellaert in #2653
• Avoid Rot2 reference-to-value returns on ARM64 by @dellaert in #2659
• Enable correct Lie-group Jacobians in Prior/Between by default by @dellaert in #2661
• Materialize Shonan certificate product by @dellaert in #2665
• Use the README infographic for the website splash by @dellaert in #2669
• Add Frobenius factor support to FastSync by @dellaert in #2671
• Prototype generic BM rotation averaging by @dellaert in #2672
• Upgrade GitHub Actions for Node 24 by @dellaert in #2674
• Optimize QCQP construction for generic BM by @dellaert in #2675
• Prefer system SuiteSparse with a namespaced CCOLAMD fallback by @ProfFan in #2676
• Add Pose2 tangent Logmap derivative overload by @dellaert in #2677
• Add gyroscope integration helpers by @dellaert in #2678
• docs: explain IMU integration covariance by @suhaslord in #2680
• Remove vendored Ceres and use system Ceres for timing targets by @ProfFan in #2681
• Refactor AHRS and IMU preintegration hierarchy by @dellaert in #2682
• Add unified ALM update policies with paper-inspired BCL by @dellaert in #2684
• Suppress GCC 16 Spectra false positive by @ProfFan in #2685
• Avoid recursive BayesTree serialization stack overflow by @dellaert in #2686
• Add NavState Lie-group IMU preintegration by @dellaert in #2687
• Use exact rotating-Earth dynamics in IMU prediction by @dellaert in #2688
• Lie-group aware differentiable spline curves (supersedes #2212) by @dellaert in #2689
• Improve batch-factor and LM optimization hot paths by @dellaert in #2691
• Add flat Gaussian-factor kernels and BAL smart/PCG benchmarks by @dellaert in #2693
• Assemble compact point-batch sparse Schur systems by @dellaert in #2694
• Add an optional CHOLMOD backend for compact BAL Schur systems by @dellaert in #2695
• Document MultifrontalSolver and the linear module by @dellaert in #2698
• Align Python wrapper declarations with C++ APIs by @dellaert in #2699
• Remove deprecated Coriolis flag usage from examples by @dellaert in #2701
• Enhance EquivariantFilter with error coordinate handling and matrix output by @arihantb2 in #2702
• Gate IMU Jacobian finite-difference step for quaternions by @dellaert in #2704
• Do not install vendored pybind11 if we use the system build by @ProfFan in #2705
• Add CUDA nonlinear optimization with cuDSS and PCG backends by @leolrg in #2706
• wrap: declare navigation factors under NoiseModelFactor by @DLuminary in #2707
• wrap: expose adaptive reorder settings and Bayes tree fill-in by @DLuminary in #2708
• wrap: complete targeted single-key calculateEstimate coverage by @DLuminary in #2709
• Generalize fixed-size Jacobian factors to arbitrary arity by @dellaert in #2710
• wrap: expose noise model introspection and Robust accessors by @DLuminary in #2711
• Make JacobianVector opaque for every module fragment by @DLuminary in #2712
• Adding certifiable estimation applications by @zhexin1904 in #2713
• Corrects authorship note in AI generated file CardinalSplineBasis.h by @BrettRD in #2717
• wrap: complete the fixed-lag smoother state accessors by @DLuminary in #2720
• Expose the remaining per-update key sets on ISAM2Result by @DLuminary in #2721
• Move CUDA SFM implementation and docs into sfm by @dellaert in #2722
• Deprecate stale gtsam_unstable APIs by @dellaert in #2724
• Add system libraries CI workflow by @dellaert in #2726
• Add public CPU and CUDA SFM solver paths by @dellaert in #2727
• Add sparse AllDiff support for hybrid inference by @dellaert in #2728
• Speed up discrete multiplication with generic TableFactor conversion by @dellaert in #2729
• Deduplicate variable layout metadata around KeyInfo by @leolrg in #2730
• Revamp MATLAB bindings and wrapper interfaces by @ProfFan in #2731
• Expose ISAM2 update params in ConcurrentIncrementalFilter by @dellaert in #2733
• Speed up SO3::Logmap by @DLuminary in #2735
• Promote discrete constraints to stable API by @dellaert in #2736
• Declare Python wrapper types before binding methods by @ProfFan in #2737
• Move misplaced headers to canonical modules by @dellaert in #2738
• Remove using namespace std from header by @B-Leslie in #2742
• Reduce duplicated linear solver internals by @dellaert in #2743
• Use Clang 14 for nightly Python build by @ProfFan in #2744
• Use hybrid TBB gathering for multifrontal parent updates by @dellaert in #2745
• Consolidate multifrontal block update plumbing by @dellaert in #2746
• Expose METIS random seed on Ordering::Metis by @SamanyuB910 in #2747
• Aggregate non-TBB leaf traversal tasks by @dellaert in #2748
• Avoid full parent scratch matrices without TBB by @dellaert in #2751
• Add g2o 3D landmark dataset support by @dellaert in #2754
• Wrap PIM backends through template interface by @dellaert in #2757
• Bring Galilean PIM in line with templated IMU preintegration by @dellaert in #2759
• Clarify certifiable localization notebook by @dellaert in #2760
• Add Python bindings for general CUDA Levenberg-Marquardt by @leolrg in #2761
• Remove duplication: Consolidate gravity magnitude resolution and reject PIMs without params by @nkhedekar in #2762
• Add ImuFactor2WithGravity: NavState-based gravity-aware IMU factor by @nkhedekar in #2763
• Optimize shared IMU prediction and factor linearization by @dellaert in #2764
• Optimize IMU measurement integration by @dellaert in #2765
• Add combined Galilean IMU factor by @dellaert in #2766
• Add wrapped API walkthrough notebooks by @dellaert in #2767
• Handle a timestamp supplied for a key with no value and no factor by @DLuminary in #2769
• Tolerate a missing timestamp in FixedLagSmoother::eraseKeyTimestampMap by @DLuminary in #2770
• Update Galilean IMU factor notebooks by @dellaert in #2771
• Add sam, sfm, and slam API walkthrough notebooks by @dellaert in #2772
• Default IMU factor errors to SE_2(3) Logmap by @dellaert in #2774
• Avoid full covariance construction in Gaussian::sigmas() by @owenbales in #2775
• Reuse matrix-vector products in PowerMethod::compute() by @owenbales in #2776
• python: ship a PEP 561 py.typed marker by @DLuminary in #2777
• CustomFactor: make the jacobian argument optional in its own type by @DLuminary in #2778
• ci: generate gtsam_unstable stubs when building wheels by @DLuminary in #2780
• Reject timestamps without corresponding values in IncrementalFixedLagSmoother by @talfeiner-mc in #2781
• Support left group actions in EquivariantFilter by @DLuminary in #2783
• Add D=1 MOSEK support to certifiable applications by @dellaert in #2784
• Reset adaptive LM rejection multiplier after accepted steps by @ProfFan in #2786
• Add Python bindings for CUDA GNC optimizers by @leolrg in #2787
• Move CUDA linear solver guide into doc directory by @dellaert in #2788
• Add Values::extract(keys) and calculateEstimate(keys) for subset extraction by @DLuminary in #2790
• Report expired pending keys in FixedLagSmoother::Result by @DLuminary in #2791
• Update wrap to have proper stub generation for C++ std::function callables by @ProfFan in #2794
• Anchor smart PCG regression camera and baseline by @dellaert in #2795
• Reduce iSAM2 update overhead and make tree statistics opt-in by @dellaert in #2800
• Share homogeneous coordinates in MOSEK SDP transcription by @dellaert in #2802
• Organize Python notebook examples by GTSAM module by @dellaert in #2803
• Prepare documentation and ROS colcon defaults for 4.3 by @dellaert in #2804
• Expose MOSEK SDP solvers in MATLAB by @dellaert in #2805

Full Changelog: 4.2.0...4.3.0

---

This discussion was created from the release GTSAM 4.3.0.