safe-control

JAX toolbox for safe robotics control: Control Barrier Function (CBF) safety filters, MPPI planning, and a drop-in safe-RL layer.

Safety-aware MPC-based RL framework

Safe Pontryagin Differentiable Programming (Safe PDP) is a new theoretical and algorithmic safe differentiable framework to solve a broad class of safety-critical learning and control tasks.

The Verifiably Safe Reinforcement Learning Framework

Official code repository for "Integrating Predictive Motion Uncertainties with Distributionally Robust Risk-Aware Control for Safe Robot Navigation in Crowds"

Model Predictive Control-based Reinforcement Learning with Control Barrier Functions

Code for the paper "Control Barriers in Bayesian Learning of System Dynamics"

Safe Control for Nonlinear Systems

Benchmark on interactive safety

[ICML2025] SafeDiffCon is a diffusion model for safe PDE Control.

Contains the code used to generate the results of the paper "Safe Control of Grid-Interfacing Inverters with Current Magnitude Limits"

Code for my bachelor thesis on "Control Barrier Functions for Learning-Based Regulation of Elastic Actuators with State Constraints".

Code of algorithms and simulations in article "Safe Distributed Control of Multi-Robot Systems with Communication Delays," IEEE TVT 2025.

Distributionally Robust Safe Control of Robotic Manipulators in Dynamic Environments

MATLAB toolbox for Barrier States (BaS) safe control and safety-embedded systems.

Safety-critical control for a tilt-rotor bicopter using High-Order Control Barrier Functions (HOCBF) and LQR, with guaranteed altitude constraint satisfaction in MATLAB/Simulink.

Python library for Barrier States (BaS) safe control and safety-embedded systems.

Official code and reproducibility package for D-SCEOS: swarm-inspired peer-to-peer control for resource-aware flexibility sharing in cyber-physical energy systems (CPES), distributed energy resources and virtual power plants; CLF/HOCBF safe control. Energy Conversion and Management: X 31 (2026) 102212. DOI: 10.1016/j.ecmx.2026.102212

Companion data and figures for: AEGIS - An Age-Aware, Evidence-Guarded Safe-Action Layer Fusing RGB-D Sensing with Vision-Language Monitoring for Collision-Aware Robotic Manipulation (Niryo NED3 Pro)