Abstract:Collaborative decision-making is an essential capability for multi-robot systems, such as connected vehicles, to collaboratively control autonomous vehicles in accident-prone scenarios. Under limited communication bandwidth, capturing comprehensive situational awareness by integrating connected agents' observation is very challenging. In this paper, we propose a novel collaborative decision-making method that efficiently and effectively integrates collaborators' representations to control the ego vehicle in accident-prone scenarios. Our approach formulates collaborative decision-making as a classification problem. We first represent sequences of raw observations as spatiotemporal graphs, which significantly reduce the package size to share among connected vehicles. Then we design a novel spatiotemporal graph neural network based on heterogeneous graph learning, which analyzes spatial and temporal connections of objects in a unified way for collaborative decision-making. We evaluate our approach using a high-fidelity simulator that considers realistic traffic, communication bandwidth, and vehicle sensing among connected autonomous vehicles. The experimental results show that our representation achieves over 100x reduction in the shared data size that meets the requirements of communication bandwidth for connected autonomous driving. In addition, our approach achieves over 30% improvements in driving safety.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is how to achieve effective collaborative decision - making under limited communication bandwidth in connected autonomous driving, especially in high - accident - prone scenarios. Specifically, the paper focuses on how to efficiently integrate the observations of multiple autonomous vehicles (especially the host vehicle and the collaborating vehicles) to improve the overall perception of the surrounding environment and thus make better decisions, such as taking braking actions to avoid traffic accidents. Due to the limited communication bandwidth between vehicles, it is unrealistic to directly share raw observation data. Meanwhile, occlusions in observations, dynamic street conditions, and complex interactions between objects also make collaborative decision - making very challenging. To address these challenges, the paper proposes a new spatio - temporal graph - based collaborative decision - making method. This method first represents the observation sequences of each vehicle as spatio - temporal graphs, where nodes encode the detected object positions, spatial edges encode the spatial distances between object pairs, and temporal edges encode the movement of the same object at different time points. Through this representation, connected vehicles can efficiently share sequential observations while overcoming communication limitations. Next, all host - vehicle - collaborating - vehicle representations are merged using the Global Navigation Satellite System (GNSS) attitude, and the collaborative decision - making problem is formalized as a classification problem. Finally, the spatial and temporal relationships of objects are simultaneously analyzed through a heterogeneous graph learning framework to generate an overall embedding of the scenario, and further predict whether the host vehicle should take braking actions to avoid accidents. The main contributions of this method are as follows: 1. A new representation method based on spatio - temporal graphs is proposed, which not only integrates the current state but also includes the historical movement of street - surface objects, achieving a more than 100 - fold reduction in packet size and meeting the communication bandwidth requirements for connected autonomous driving. 2. A new spatio - temporal graph neural network based on heterogeneous graph learning is designed. By analyzing the spatial and temporal connections of objects in a unified manner, it encodes cues for collaborative decision - making and improves driving safety by more than 30%. The paper verifies the effectiveness of this method through simulation tests, and the results show that it performs excellently in reducing the amount of shared data and improving driving safety.

Collaborative Decision-Making Using Spatiotemporal Graphs in Connected Autonomy

Multilevel Graph Reinforcement Learning for Consistent Cognitive Decision-making in Heterogeneous Mixed Autonomy

GSC: A Graph and Spatio-Temporal Continuity Based Framework for Accident Anticipation

Multi-Agent Learning Empowered Collaborative Decision for Autonomous Driving Vehicles

Decision-Making for Autonomous Vehicles with Interaction-Aware Behavioral Prediction and Social-Attention Neural Network

SIF-STGDAN: A Social Interaction Force Spatial-Temporal Graph Dynamic Attention Network for Decision-Making of Connected and Autonomous Vehicles

Multi-Granularity Collaborative Decision with Cognitive Networking in Intelligent Transportation Systems

Asynchronous Collaborative Localization by Integrating Spatiotemporal Graph Learning with Model-Based Estimation

Graph neural network and reinforcement learning for multi‐agent cooperative control of connected autonomous vehicles

GITSR: Graph Interaction Transformer-based Scene Representation for Multi Vehicle Collaborative Decision-making

Modeling spatio-temporal interactions for vehicle trajectory prediction based on graph representation learning.

Cooperative Decision-Making of Connected and Autonomous Vehicles in an Emergency

Collaborative Perception for Connected and Autonomous Driving: Challenges, Possible Solutions and Opportunities

Graph Convolution-Based Deep Reinforcement Learning for Multi-Agent Decision-Making in Interactive Traffic Scenarios.

Spatiotemporal Scene-Graph Embedding for Autonomous Vehicle Collision Prediction

Spatio-Temporal Scene-Graph Embedding for Autonomous Vehicle Collision Prediction

Spatiotemporal Learning of Multivehicle Interaction Patterns in Lane-Change Scenarios

An End-to-End Collaborative Learning Approach for Connected Autonomous Vehicles in Occluded Scenarios

Multi-agent Driven Collaborative Decision Mechanism of Information Fusion for Autonomous Driving Vehicles

InterCoop: Spatio-Temporal Interaction Aware Cooperative Perception for Networked Vehicles

A Universal Cooperative Decision-Making Framework for Connected Autonomous Vehicles with Generic Road Topologies