Multi-Task End-to-End Self-Driving Architecture for CAV Platoons

Abstract

Connected and autonomous vehicles (CAVs) could reduce emissions, increase road safety, and enhance ride comfort. Multiple CAVs can form a CAV platoon with a close inter-vehicle distance, which can further improve energy efficiency, save space, and reduce travel time. To date, there have been few detailed studies of self-driving algorithms for CAV platoons in urban areas. In this paper, we therefore propose a self-driving architecture combining the sensing, planning, and control for CAV platoons in an end-to-end fashion. Our multi-task model can switch between two tasks to drive either the leading or following vehicle in the platoon. The architecture is based on an end-to-end deep learning approach and predicts the control commands, i.e., steering and throttle/brake, with a single neural network. The inputs for this network are images from a front-facing camera, enhanced by information transmitted via vehicle-to-vehicle (V2V) communication. The model is trained with data captured in a simulated urban environment with dynamic traffic. We compare our approach with different concepts used in the state-of-the-art end-to-end self-driving research, such as the implementation of recurrent neural networks or transfer learning. Experiments in the simulation were conducted to test the model in different urban environments. A CAV platoon consisting of two vehicles, each controlled by an instance of the network, completed on average 67% of the predefined point-to-point routes in the training environment and 40% in a never-seen-before environment. Using V2V communication, our approach eliminates casual confusion for the following vehicle, which is a known limitation of end-to-end self-driving.

Advanced Sensing and Control for Connected and Automated Vehicles

“…This Special Issue of Sensors aims at reporting on some of the recent research efforts on this increasingly important topic. The 12 accepted papers in this Issue cover vehicle position estimation [1], vehicle dynamic parameters estimation [2], cooperative collision warning systems [3], small object detection [4], impact identification of the driver's driving performance on executive control function [5], hybrid path planning for autonomous driving [6], trajectory tracking for autonomous driving [7], vehicle stability control [8], vehicle stability and ride comfort control [9], urban platooning protocol design for platoon [10], path planning algorithm for platooning [11], and self-driving architecture design for CAV platoon [12].…”

“…A self-driving architecture combining the sensing, planning, and control for CAV platoons in an end-to-end fashion is proposed in [12]. This multi-task model can switch between two tasks to drive either the leading or following vehicle in the platoon.…”

“…In summary, there is a huge potential for CAV in collision avoidance, safety improvement and driving stability improvement. The papers gathered in this Special Issue contributed by proposing solutions to the general problem of state and/or parameter estimation [1][2][3], obstacle detection [4], driver behavior estimation [5], and/or classical academic problems such as path planning [6], path tracking [7], and stability control for autonomous driving [8,9] and/or by suggesting applications of the combined use of sensors and advanced algorithms to platooning [10][11][12], thus showing the theoretical challenges and practical interest of this research topic. Finally, we wish to thank the authors, reviewers, and journal staff for their commitment and effort, which made it possible to complete this Special Issue on time.…”

A Convolutional Neural Network-Based End-to-End Self-Driving Using LiDAR and Camera Fusion: Analysis Perspectives in a Real-World Environment

“…Navarro et al proposed sensor fusion-based E2E self-driving using real-world acquired data [15], but they did not analyze how to work their algorithm in real-world situations such as an urban traffic signal. Huch et al suggested V2X-based E2E self-driving for platooning [16]. Prashanth et al proposed JacintoNet, which was implemented in Texas Instruments (TI) TDA2x System on Chip, for real-time working [17].…”

Load Effect of Automated Truck Platooning on Highway Bridges and Loading Strategy

“…Actually, the mandatory safe distances for ordinary vehicles are generally no less than 50 m. For heavy-duty trucks, the requirement is usually longer than 100 m [ 13 ]. Promoted by the advanced 5G, V2X communication technologies [ 14 ], and autonomous driving technology [ 15 ], the following gap between vehicles can be shortened to less than 15 m [ 16 ]. Consequently, the transportation density will be significantly increased.…”