Multi-Uncertainty Aware Autonomous Cooperative Planning

TL;DR

A novel multi-uncertainty aware ACP framework that simultaneously accounts for multiple types of uncertainties via regularized cooperative model predictive control (RC-MPC) is proposed.

Abstract

Autonomous cooperative planning (ACP) is a promising technique to improve the efficiency and safety of multi-vehicle interactions for future intelligent transportation systems. However, realizing robust ACP is a challenge due to the aggregation of perception, motion, and communication uncertainties. This paper proposes a novel multi-uncertainty aware ACP (MUACP) framework that simultaneously accounts for multiple types of uncertainties via regularized cooperative model predictive control (RC-MPC). The regularizers and constraints for perception, motion, and communication are constructed according to the confidence levels, weather conditions, and outage probabilities, respectively. The effectiveness of the proposed method is evaluated in the Car Learning to Act (CARLA) simulation platform. Results demonstrate that the proposed MUACP efficiently performs cooperative formation in real time and outperforms other benchmark approaches in various scenarios under imperfect knowledge of the environment.

Figures (8)

• Figure 1: Perception uncertainty and multi-vehicle perception.
• Figure 2: System architecture of MUACP, which integrates multiple uncertainties and vehicle cooperative planning.
• Figure 3: Proposed lane-change motion with (a) 3 AVs. (b) 4 AVs. (c) 5 AVs. (d) 6 AVs.
• Figure 4: The state profile of proposed lane-change motion with 3 AVs in normal weather condition: (a) motion accelerations, (b) motion steering angles; and rainy weather condition: (c) motion accelerations, (d) motion steering angles.
• Figure 5: The 6-AV case at bi-directional traffic road under no uncertainty: (a) State and control profiles; (b) 6 AV motion accelerations; (c) 6 AV motion steering angles.