This work studies the design of safe control policies for large-scale non-linear systems operating in uncertain environments. In such a case, the robust control framework is a principled approach to safety that aims to maximize the worst-case performance of a system. However, the resulting optimization problem is generally intractable for non-linear systems with continuous states. To overcome this issue, we introduce two tractable methods that are based either on sampling or on a conservative approximation of the robust objective. The proposed approaches are applied to the problem of autonomous driving.


Reproduce the experiments

Install requirements

Run the benchmark

cd <path-to-rl-agents>/scripts/
python experiments.py benchmark configs/RoundaboutEnv/benchmark_robust_control.json \
                      --test --episodes=100 --processes=4

The following agents will be evaluated:

    "environments": [
    "agents": [