Paper Info
Title
A Novel Direct Trajectory Planning Approach Based on Generative Adversarial Networks and Rapidly-Exploring Random Tree
Abstract
Trajectory planning is essential for self-driving vehicles and has stringent requirements for accuracy and efficiency. The existing trajectory planning methods have limitations in the feasibility of planned trajectories and computational efficiency. This paper proposes a life-long learning framework to achieve effective and high-accuracy direct trajectory planning (DTP) tasks. Based on generative adversarial networks (GANs), this study develops a lightweight GDTP model to map the initial/final states and the control action sequence. Additionally, by embedding the GDTP into the rapidly-exploring random tree (RRT), a GDTP-RRT algorithm is further designed for long-distance and multi-stage planning tasks. Taking the tractor-trailer as an application case, we test the proposed method in multiple scenarios with varying characteristics. The experimental results show that the method can plan highly feasible trajectories in a short time, compared with the most applied algorithm – the cubic curve RRT* (CCRRT*). It is found that the tracking errors of our method are 29.1% and 44.1% lower than the CCRRT* in terms of position and heading angle. This paper provides an effective and stable vehicle trajectory planning method for complex self-driving tasks.
Year
DOI
Venue
2022
10.1109/TITS.2022.3164391
IEEE Transactions on Intelligent Transportation Systems
Keywords
DocType
Volume
Self-driving vehicles,trajectory planning,generative adversarial networks,rapidly-exploring random tree,tractor-trailer
Journal
23
Issue
ISSN
Citations 
10
1524-9050
0
PageRank 
References 
Authors
0.34
15
5
Name
Order
Citations
PageRank
Cong Zhao100.34
Yifan Zhu200.34
Yuchuan Du343.21
Feixiong Liao400.34
Ching-Yao Chan57923.48