Implementation of SE3-Transformers for Equivariant Self-Attention, in Pytorch. This specific repository is geared towards integration with eventual Alphafold2 replication.
-
Updated
Aug 28, 2025 - Python
Implementation of SE3-Transformers for Equivariant Self-Attention, in Pytorch. This specific repository is geared towards integration with eventual Alphafold2 replication.
An Open-source Strong Baseline for SE(3) Planning in Autonomous Drone Racing
Implementation of Lie Transformer, Equivariant Self-Attention, in Pytorch
A complete, hardware-ready Python package for Koopman-based Linear Model Predictive Control (LMPC), delivering real-time trajectory tracking for quadrotors using analytical Koopman lifting (no training data required)
Robot Motion Estimate: Tools, Variables, and Factors for SLAM in robotics; also see Caesar.jl.
SE3 interpolation and Quat+R^3 interpolation are implemented.
Lie groups and algebra with some quaternions
Register Gaussian splats: align & merge two 3DGS scans into one SE(3)/Sim(3) frame — classical, learned, or zero-shot (BUFFER-X) seeds + splat-native refine. pip install splatreg.
A C++ Eigen-based maths library for robotics applications, including SO(3) and SE(3) transformations, screw motions, robot kinematics, and control laws. All the codes are successfully tested in ROS 2 humble.
Numerically stable implementation of batched SE(3) exponential and logarithmic maps
Companion code for arXiv:2605.02252 — exact higher-order derivatives for SE(3) NLL via hybrid analytical/AD methods.
Benchmarking library for SE(3) Manifold Optimization: Geometric Dual Quaternions (GeoDQ), ESKF, and UKF-M implementations for Robotics & Navigation
Lightweight C++ header-only template library for translation, rotation and homogeneous transformation. Requires C++17 or Later. No dependencies with other libraries and stl.
Modern C++20 motion core for industrial robotics, targeting kinematics, dynamics, trajectory generation, calibration, and robust numerics.
SE(3) geodesic action head for VLA models
Pose-graph SLAM back-end written from scratch: SE(2)/SE(3) Lie group Jacobians, sparse Levenberg-Marquardt with variable reordering, and robust kernels. Evaluated on standard public benchmark datasets.
GAUSS-VLA: certified safe VLA via SE(3) flow matching
To associate your repository with the se3 topic, visit your repo's landing page and select "manage topics."