Don't Get Stuck: A Deadlock Recovery Approach

TL;DR

This paper presents a hybrid planning and control framework combining hybrid-A*, STL, and MPPI to recover from deadlocks in multi-agent navigation, ensuring safety and compliance in complex traffic scenarios.

Contribution

It introduces a novel STL-MPPI framework that integrates path planning, rule-based constraints, and dynamic refinement for deadlock recovery in shared-space navigation.

Findings

Effective deadlock recovery demonstrated in simulations.

Validated safety and rule compliance on scaled cars.

Framework adaptable to complex traffic environments.

Abstract

When multiple agents share space, interactions can lead to deadlocks, where no agent can advance towards its goal. This paper addresses this challenge with a deadlock recovery strategy. In particular, the proposed algorithm integrates hybrid-A$^\star$, STL, and MPPI frameworks. Specifically, hybrid-A$^\star$ generates a reference path, STL defines a goal (deadlock avoidance) and associated constraints (w.r.t. traffic rules), and MPPI refines the path and speed accordingly. This STL-MPPI framework ensures system compliance to specifications and dynamics while ensuring the safety of the resulting maneuvers, indicating a strong potential for application to complex traffic scenarios (and rules) in practice. Validation studies are conducted in simulations and on scaled cars, respectively, to demonstrate the effectiveness of the proposed algorithm.