On String Stability of Mixed Autonomous and Human-driven Vehicle Platoons with Transmissibility-based Health Monitoring

The safe coexistence of connected autonomous vehicle platoons with the current on-road human-driven vehicles is a challenging issue since human-driver behavior is unknown and difficult to estimate. Different physical and cyber faults in CAV platoons were mitigated using transmissibility-based sliding mode control in previous work. Transmissibility-based health monitoring uses available sensor measurements only and does not require knowledge of the excitation or the dynamics of the platoon. We consider a mixed autonomous and human-driven platoon, with the human-driver behavior as an independent excitation signal. Therefore, the transmissibility-based health monitoring technique is independent of the desired velocity, human-driver behavior, and the underlying dynamics of the platoon. In this paper, we investigate the stability of the transmissibility-based sliding mode controller used for fault mitigation. Next, we investigate the string stability of the faulty mixed autonomous and human-driven platoon, with the fault being mitigated using transmissibility-based sliding mode control. We show that if the sliding mode controller stability is attained, then the faulty platoon with mitigated fault is always string stable as well.

I. Introduction

The interest in connected autonomous vehicles (CAV) evolved in the past decades as CAV platoons can enhance traffic safety, ease congestion, and positively impact the environment through autonomous platoon control [1]. The evolution of reaching a higher level of autonomy in CAV platoons was accomplished by developing two aspects. The first aspect is developing perception sensors such as radar and LiDAR, which allow autonomous vehicles to supervise their surroundings. The second aspect is developing the cyber components of the platoon, such as vehicle-to-vehicle (V2V) communication links to cover longer ranges and higher rates of data transmission [2]. However, merging this new technology with the current on-road human-driven vehicles should be addressed in terms of safety, security, and stability.