We are pleased to release GTSAM 3.2.0. This is a maintenance release with no major API changes. Notable changes since the 3.1.0 release include:

• Full support for Boost 1.56 across all tested platforms. Visual Studio 2013 is now supported with Boost 1.56 or higher.
• Improved rotation Logmap accuracy in quaternion mode.
• Improved support for the latest version of the Intel Math Kernel Library. CMake find scripts were updated to search additional installation paths.
• Initialization approaches for 3D pose graphs (paper submitted to ICRA, more details soon.)
• Lots of other small bugfixes and improvements

Information and Download

We welcome your feedback and questions at gtsam@lists.gatech.edu.

We are pleased to release GTSAM 3.0.0, a major update from the previous 2.3.1 release.  The primary improvements in GTSAM 3.0.0 result from a complete rewrite of the core linear and symbolic optimization pipeline. Multithreading using TBB is now mostly ubiquitous, and redesigning algorithms eliminated much bookkeeping, both resulting in significant performance improvements and data structures that are easier to deal with in user code.

 Details:

• Complete rewrite of the linear and symbolic solver with the following improvements:
  • Multithreaded from the ground up using Intel TBB.
  • Numerous performance improvements by redesign that eliminates much bookkeeping.
• Computationally-intensive code now parallelized with TBB (in addition to the core solver).
• MKL automatically detected and used for further performance improvements.
• Performance and functionality improvements in ISAM2.

Information and DownloadDetailed Changelist

We welcome your feedback and questions at gtsam@lists.gatech.edu.

We are pleased to announce GTSAM 2.3.0!  Notable changes from 2.2.0 include:

• Added a basic IMU (inertial navigation) factor for estimating 6-dof pose, translational velocity, and IMU/gyro bias (supports both time-varying and constant bias).
• Added an "equivalent summarized" IMU factor with the same capabilities as the basic one, but that precomputes a summary of a large number of IMU measurements to allow for efficient relinearization (following Lupton and Sukkarieh in TRO 2012).  Currently uses numerical derivatives, but an update to be released shortly as part of GTSAM 2.3.1 will use analytic derivatives for improved computational efficiency.
• New feature - basic "summarization" of nonlinear systems - Given an existing nonlinear graph and a linearization point, summarization creates a new nonlinear system on a specified subset of variables in the graph. The summarization is obtained by marginalizing out other variables in the linearized system. This new summarized graph can be added back into a nonlinear graph under a linearization assumption.
• Major speed improvements in iSAM2 when dealing with many, >50,000, variables (e.g. very long chains).
• Several build system improvements (see detailed changelist)

Information and DownloadDetailed Changelist

We welcome your feedback and questions at gtsam@lists.gatech.edu.

Alongside the release of GTSAM 2.1.0 (see below) is a new hands-on tutorial for learning about factor graphs and GTSAM.  This tutorial will help you to quickly get started writing your own software using GTSAM as a factor graph optimization backend.  Also, it provides a very accessible introduction, to the mathematics and applications of factor graphs.  The tutorial covers probability functions represented by factor graphs and their optimization, a number of real-world mapping examples with source code, and how to easily have GTSAM optimize your own custom factors.

Factor Graphs and GTSAM: A Hands-on Introduction, Frank Dellaert, Technical Report GT-RIM-CP&R-2012-002.