Demonstration for STFC method

STFC here refers to our proposed spatio-temporal formation control method.

Jianghong Dong1, Jiawei Wang2, Mengchi Cai1, Yibin Yang1, Qing Xu1, Jianqiang Wang1 and Keqiang Li1
1Tsinghua University, 2University of Michigan

Experimental scenario


Descriptive Alt Text

Three sequential experimental scenarios encountered during continuous formation motion simulation:

1) Scenario A: A small obstacle is present in the uppermost lane of the formation.

2) Scenario B: An obstacle vehicle is present in the lowermost lane of the formation.

3) Scenario C: The formation passes the obstacle vehicle.

Continuous formation motion simulation


Initially, the formation encounters a small obstacle within the uppermost lane. The vehicles in the uppermost lane assess that they can handle the obstacle individually and proceed to avoid it through intra-lane maneuvers while maintaining good formation performance. Subsequently, the formation encounters a slow-moving obstacle vehicle ahead and triggers a formation transition. Finally, to fully utilize the road resources, the vehicles smoothly transition back from a two-lane formation to a three-lane formation.

Abstract

Formation control of Connected and Autonomous Vehicles (CAVs) has shown significant potential for improving traffic safety and efficiency in multi-lane traffic. However, previous work has primarily focused on spatial coordination without temporal considerations, which significantly limits their practical applicability in real-world traffic. In this paper, we propose a Spatio-Temporal Formation Control method (STFC) that integrates centralized formation generation with distributed trajectory planning. Precisely, we propose a graph-based formation maintenance representation, and show that the interlaced geometric structure is optimal for multi-lane formation. Then, we develop a distributed spatio-temporal joint formation trajectory planning method that simultaneously optimizes spatial positions and temporal duration, with consideration of multiple objectives such as formation maintenance and obstacle avoidance. Further, we design a polynomial vehicle-to-target assignment algorithm that inherently resolves conflicts. Simulation experiments demonstrate the superiority of our approach over baseline methods in terms of formation maintenance and transition.

Method framework


Descriptive Alt Text