GTSAM is a sensor fusion library based on factor graphs

Almost at 4.3!!!
GTSAM is going GPU! 🚀 Ruogu Li has built an impressive new experimental CUDA backend for nonlinear factor-graph optimization in GTSAM, just in time for the 4.3 release next week. Read our blog post: gtsam.org/2026/08/20/cuda-ba… #GTSAM #CUDA #Robotics #SLAM #StructureFromMotion #BundleAdjustment #ComputerVision #GPU #OpenSource
2
241
GTSAM 4.3 retweeted
Custom factors are one of the fastest ways to prototype new residuals in GTSAM, and they now work across Python, C++, and MATLAB. Here is an example with localization from known landmarks:
1
3
44
3,018
GTSAM 4.3 retweeted
Marginal covariance recovery has always been fast in @gtsam4 , as it exploits the structure of Bayes trees. Now, thanks to Steiner subtrees, it will be super fast for arbitrary covariance queries.
1
6
44
2,873
GTSAM 4.3 retweeted
As an application of the invariant EKF in GTSAM, we added a state estimator for legged robots, along with some more factor-graph-heavy variants. Try it on your rosbags - especially interested in validating it with non-quadrupeds (bipeds, hexapods, spiders). It takes IMU and feet in body frame as inputs.
1
10
75
4,908
GTSAM 4.3 retweeted
We just released the first alpha version of GTSAM 4.3. It has been a long time coming, but happy to share our many updates during #ICRA2025 . Release link below. In Python: “pip install gtsam==4.3a0”
2
13
84
9,286
Trajectory optimization and control for quad rotors in GTSAM. Runs in colab, as well!
Replying to @sethahutchinson
@sethahutchinson and I uploaded a 101 intro to quad-rotor trajectory optimization and control at roboticsbook.org/S75_drone_p… . - This draft is targeted at undergrads, let me know what you think! - There is an “open in Colab” button, so you can run the examples right in your browser!
1
4
1,329
GTSAM 4.3 retweeted
After GTSAM, GTSfM, and GTDynamics, you just *know* GTGPT is next, right ? :-) Here is a GPT for interacting with factor graphs: chat.openai.com/g/g-wgqbXvcQ…
2
9
67
10,673
A lot of Georgia Tech papers at IROS 2022 in Kyoto, including a few GTSAM-heavy ones: research.gatech.edu/robotics…
1
1
1
Deep learning and @gtsam4 . What’s not to like?
Excited to share our work arxiv.org/abs/2108.02274 at #CoRL2021 on learning observation models end-to-end in factor graphs! We show that learning observation models can be viewed as shaping energy functions that graph optimizers, even non-differentiable ones, optimize.🧵 1/
4
GTSAM 4.3 retweeted
⁦@gtsam4⁩ rendered as colorful robot graffiti by ⁦@gchenfc⁩. Any questions?
2
13