End-to-End Autonomous Driving Decision Based on Deep Reinforcement Learning

被引：0
|
作者：
Huang Z.-Q. [1 ,3 ]
Qu Z.-W. [1 ,3 ]
Zhang J. [1 ,3 ]
Zhang Y.-X. [2 ]
Tian R. [1 ,3 ]
机构：
[1] Faculty of Information Technology, Beijing University of Technology, Beijing
[2] School of Electronic and Information Engineering, Beijing Jiaotong University, Beijing
[3] Beijing Engineering Research Center for IoT Software and Systems, Beijing
来源：
Tien Tzu Hsueh Pao/Acta Electronica Sinica | 2020年 / 48卷 / 09期
关键词：
Autonomous driving; DDPG; Deep reinforcement learning; End-to-end decision-making;
D O I：
10.3969/j.issn.0372-2112.2020.09.007
中图分类号：
学科分类号：
摘要：
The end-to-end driving decision making is a research hotspot in the field of autonomous driving. This paper studies the end-to-end driving decision of continuous action output based on DDPG (Deep Deterministic Policy Gradient) deep reinforcement learning algorithm. First, an end-to-end decision-making control model based on DDPG algorithm is established. The model outputs the continuous control quantity of vehicle driving action (acceleration, braking, steering) according to the continuously acquired perception information (such as vehicle angle, vehicle speed, road distance, etc. ) as the input state. Then, the model is trained and verified in different driving environments on the platform of TORCS (The Open Racing Car Simulator). The results show that the model can realize the end-to-end decision-making of autonomous driving. At last, it is compared with DQN(Deep Q-Learning Network) model of discrete action output. The experimental results show that DDPG model has better decision control effect. © 2020, Chinese Institute of Electronics. All right reserved.
引用
收藏
页码：1711 / 1719
页数：8
相关论文
共 22 条
  • [1] XIONG Lu, KANG Yu-chen, ZHANG Pei-zhi, Et al., Research on behavior decision-making system for unmanned vehicle, Automobile Technology, 515, pp. 1-9, (2018)
  • [2] LIU Guo-rong, ZHANG Yang-ming, Trajectory tracking of mobile robots based on fuzzy PID-P type iterative learning control, Acta Electronica Sinica, 41, 8, pp. 1536-1541, (2013)
  • [3] POMERLEAU D A., ALVINN: An autonomous land vehicle in a neural network, Advances in Neural Information Processing Systems, pp. 305-313, (1989)
  • [4] MULLER U, BEN J, COSATTO E, Et al., Off-road obstacle avoidance through end-to-end learning, Advances in Neural Information Processing Systems, pp. 739-746, (2006)
  • [5] BOJARSKI M, DEL TESTA D, DWORAKOWSKI D, Et al., End to End Learning for Self-driving Cars, (2016)
  • [6] BOJARSKI M, YERES P, CHOROMANSKA A, Et al., Explaining How a Deep Neural Network Trained with End-to-end Learning Steers a Car, (2017)
  • [7] WANG X, JIANG R, LI L, Et al., Capturing car-following behaviors by deep learning, IEEE Transactions on Intelligent Transportation Systems, 19, 3, pp. 910-920, (2018)
  • [8] XU H Z, GAO Y, YU F, Et al., End-to-end learning of driving models from large-scale video datasets, 2017 IEEE Conference on Computer Vision and Pattern Recognition, pp. 3530-3538, (2017)
  • [9] ZHANG X, SUN J, QI X, Et al., Simultaneous modeling of car-following and lane-changing behaviors using deep learning, Transportation Research, 104, pp. 287-304, (2019)
  • [10] LOIACONO D, PRETE A, LANZI P L, Et al., Learning to overtake in torcs using simple reinforcement learning, Proceedings of the IEEE Congress on Evolutionary Computation, pp. 1-8, (2010)
123