ROLGSYApr 15, 2022

Safe Reinforcement Learning Using Black-Box Reachability Analysis

Georgia Tech
arXiv:2204.07417v242 citationsh-index: 97
Originality Incremental advance
AI Analysis

This addresses safety concerns for robot deployment in uncertain environments, though it appears incremental as it builds on existing safe RL approaches.

The paper tackled the problem of ensuring safety in reinforcement learning for robots without known models, proposing a black-box reachability-based safety layer that outperformed other state-of-the-art methods in simulations across multiple robot types.

Reinforcement learning (RL) is capable of sophisticated motion planning and control for robots in uncertain environments. However, state-of-the-art deep RL approaches typically lack safety guarantees, especially when the robot and environment models are unknown. To justify widespread deployment, robots must respect safety constraints without sacrificing performance. Thus, we propose a Black-box Reachability-based Safety Layer (BRSL) with three main components: (1) data-driven reachability analysis for a black-box robot model, (2) a trajectory rollout planner that predicts future actions and observations using an ensemble of neural networks trained online, and (3) a differentiable polytope collision check between the reachable set and obstacles that enables correcting unsafe actions. In simulation, BRSL outperforms other state-of-the-art safe RL methods on a Turtlebot 3, a quadrotor, a trajectory-tracking point mass, and a hexarotor in wind with an unsafe set adjacent to the area of highest reward.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes