Collaborative navigation becomes essential in situations of occluded scenarios in autonomous driving where independent driving policies are likely to lead to collisions. One promising approach to address this issue is through the use of Vehicle-to-Vehicle (V2V) networks that allow for the sharing of perception information with nearby agents, preventing catastrophic accidents. In this article, we propose a collaborative control method based on a V2V network for sharing compressed LiDAR features and employing Proximal Policy Optimisation to train safe and efficient navigation policies. Unlike previous approaches that rely on expert data (behaviour cloning), our proposed approach learns the multi-agent policies directly from experience in the occluded environment, while effectively meeting bandwidth limitations. The proposed method first prepossesses LiDAR point cloud data to obtain meaningful features through a convolutional neural network and then shares them with nearby CAVs to alert for potentially dangerous situations. To evaluate the proposed method, we developed an occluded intersection gym environment based on the CARLA autonomous driving simulator, allowing real-time data sharing among agents. Our experimental results demonstrate the consistent superiority of our collaborative control method over an independent reinforcement learning method and a cooperative early fusion method.

翻译：在自动驾驶的遮挡场景中，独立驾驶策略极易导致碰撞，协同导航因此变得至关重要。解决此问题的一个有效途径是利用车对车（V2V）网络，该网络允许与邻近智能体共享感知信息，从而避免灾难性事故。本文提出一种基于V2V网络的协同控制方法，用于共享压缩的激光雷达特征，并采用近端策略优化算法来训练安全高效的导航策略。与以往依赖专家数据（行为克隆）的方法不同，所提方法直接在遮挡环境中的经验中学习多智能体策略，同时有效满足带宽限制。该方法首先对激光雷达点云数据进行预处理，通过卷积神经网络（CNN）获取有意义的特征，随后将其与邻近的网联自动驾驶车辆（CAV）共享，以警示潜在危险情况。为评估所提方法，我们基于CARLA自动驾驶模拟器开发了一个遮挡交叉路口仿真环境，支持智能体间的实时数据共享。实验结果表明，我们的协同控制方法在性能上持续优于独立强化学习方法与协同早期融合方法。