← Back ICRA 2026

Unveiling Uncertainty-Aware Autonomous Cooperative Learning Based Planning Strategy

Shiyao Zhang, Liwei Deng, Shuyu Zhang, Weijie Yuan, Hong Zhang

PDF

AI summary

Key figure (auto-extracted from paper)

A GRU-enhanced deep reinforcement learning framework effectively handles simultaneous perception and communication uncertainties to enable safe, cooperative multi-vehicle planning.

Autonomous Cooperative Planning Deep Reinforcement Learning Uncertainty Awareness GRU-SAC Collision Avoidance CARLA Simulation

Problem

Existing autonomous cooperative planning strategies struggle to simultaneously address perception, planning, and communication uncertainties, leading to degraded safety and performance under imperfect multi-vehicle information.

Approach

The authors introduce a Deep RL-based Autonomous Cooperative Planning (DRLACP) framework that uses a GRU-enhanced Soft Actor-Critic algorithm to learn collision-free trajectories while explicitly modeling LiDAR errors and V2V communication outages.

Key results

Jointly models perception and communication uncertainties within a unified cooperative planning framework
Introduces a GRU-enhanced Soft Actor-Critic algorithm to capture temporal dependencies in stochastic driving environments
Derives a probabilistic collision avoidance model that dynamically adjusts safety distances based on uncertainty confidence scores
Demonstrates superior planning performance and robustness across multiple CARLA simulation scenarios compared to baseline methods

Why it matters

Provides a robust, uncertainty-aware planning solution critical for deploying safe multi-vehicle autonomous systems in real-world environments with imperfect sensors and connectivity.

Abstract

In future intelligent transportation systems, au- tonomous cooperative planning (ACP), becomes a promising tech- nique to increase the effectiveness and security of multi-vehicle interactions. However, multiple uncertainties cannot be fully addressed for existing ACP strategies, e.g. perception, planning, and communication uncertainties. To address these, a novel deep reinforcement learning-based autonomous cooperative planning (DRLACP) framework is proposed to tackle various uncertainties on cooperative motion planning schemes. Specifically, the soft actor-critic (SAC) with the implementation of gate recurrent units (GRUs) is adopted to learn the deterministic optimal time-varying actions with imperfect state information occurred by planning, communication, and perception uncertainties. In addition, the real-time actions of autonomous vehicles (AVs) are demonstrated via the Car Learning to Act (CARLA) simulation platform. Evaluation results show that the proposed DRLACP learns and performs cooperative planning effectively, which outperforms other baseline methods under different scenarios with imperfect AV state information.

Index terms

Collision Avoidance Motion and Path Planning Reinforcement Learning