Pinned Loading

1. SARRL SARRL Public

   Safe adaptive residual reinforcement learning for robotic manipulation under model mismatch, combining physics-based control, Soft Actor-Critic, uncertainty-aware residual dynamics, and hard HOCBF …

   Python

2. ppo-robot-controller ppo-robot-controller Public

   From-scratch PPO (PyTorch) on a PyBullet reaching task - GAE bootstrap verified against a hand-derived reference

   Python

3. stackbound stackbound Public

   Static worst-case stack analyser for Cortex-M firmware / models interrupt preemption, resolves indirect calls, every bound checked against QEMU

   Python 1

4. photon-counting-lidar photon-counting-lidar Public

   Depth estimation from single photon arrival times: SPAD detector model, three estimators, and the Cramer Rao bound they are measured against. C99, no allocation.

   C

5. residual-collision-detection residual-collision-detection Public

   Sensorless collision detection for robot arms / momentum-based residual observer, verified against Pinocchio

   C 1

6. qcsi qcsi Public

   Wi-Fi CSI sensing on a microcontroller - fixed-point phase recovery, features and classification in C99

   C 2