Planning on a (Risk) Budget: Safe Non-Conservative Planning in Probabilistic Dynamic Environments

TL;DR

This paper introduces a risk-aware planning algorithm that guarantees safety bounds while avoiding overly conservative behavior in dynamic environments with uncertain agents, demonstrated in autonomous driving scenarios.

Contribution

It proposes a novel receding horizon algorithm that maintains a dynamic risk budget and guarantees safety bounds through interval risk bounds (IRBs).

Findings

Safer than baseline methods in simulated driving scenarios.

Less conservative while maintaining safety guarantees.

Successfully tested on an autonomous class 8 truck.

Abstract

Planning in environments with other agents whose future actions are uncertain often requires compromise between safety and performance. Here our goal is to design efficient planning algorithms with guaranteed bounds on the probability of safety violation, which nonetheless achieve non-conservative performance. To quantify a system's risk, we define a natural criterion called interval risk bounds (IRBs), which provide a parametric upper bound on the probability of safety violation over a given time interval or task. We present a novel receding horizon algorithm, and prove that it can satisfy a desired IRB. Our algorithm maintains a dynamic risk budget which constrains the allowable risk at each iteration, and guarantees recursive feasibility by requiring a safe set to be reachable by a contingency plan within the budget. We empirically demonstrate that our algorithm is both safer and less conservative than strong baselines in two simulated autonomous driving experiments in scenarios involving collision avoidance with other vehicles, and additionally demonstrate our algorithm running on an autonomous class 8 truck.