FAILED
首站-论文投稿智能助手
典型文献
Decision-Making Models Based on Meta-Reinforcement Learning for Intelligent Vehicles at Urban Intersections
文献摘要:
Behavioral decision-making at urban intersections is one of the primary difficulties cur-rently impeding the development of intelligent vehicle technology. The problem is that existing decision-making algorithms cannot effectively deal with complex random scenarios at urban inter-sections. To deal with this, a deep deterministic policy gradient (DDPG) decision-making algo-rithm (T-DDPG) based on a time-series Markov decision process (T-MDP) was developed, where the state was extended to collect observations from several consecutive frames. Experiments found that T-DDPG performed better in terms of convergence and generalizability in complex intersec-tion scenarios than a traditional DDPG algorithm. Furthermore, model-agnostic meta-learning (MAML) was incorporated into the T-DDPG algorithm to improve the training method, leading to a decision algorithm (T-MAML-DDPG) based on a secondary gradient. Simulation experiments of intersection scenarios were carried out on the Gym-Carla platform to verify and compare the deci-sion models. The results showed that T-MAML-DDPG was able to easily deal with the random states of complex intersection scenarios, which could improve traffic safety and efficiency. The above decision-making models based on meta-reinforcement learning are significant for enhancing the decision-making ability of intelligent vehicles at urban intersections.
文献关键词:
作者姓名:
Xuemei Chen;Jiahe Liu;Zijia Wang;Xintong Han;Yufan Sun;Xuelong Zheng
作者机构:
School of Mechanical Engineering,Beijing Insti-tute of Technology,Beijing 100081,China
引用格式:
[1]Xuemei Chen;Jiahe Liu;Zijia Wang;Xintong Han;Yufan Sun;Xuelong Zheng-.Decision-Making Models Based on Meta-Reinforcement Learning for Intelligent Vehicles at Urban Intersections)[J].北京理工大学学报(英文版),2022(04):327-339
A类:
Intersections,intersec
B类:
Decision,Making,Models,Based,Reinforcement,Learning,Intelligent,Vehicles,Urban,Behavioral,decision,making,urban,intersections,one,primary,difficulties,cur,rently,impeding,development,intelligent,technology,problem,that,existing,algorithms,cannot,effectively,deal,complex,random,scenarios,To,this,deep,deterministic,policy,gradient,DDPG,series,Markov,process,MDP,was,developed,where,extended,collect,observations,from,several,consecutive,frames,Experiments,found,performed,better,terms,convergence,generalizability,than,traditional,Furthermore,agnostic,meta,learning,MAML,incorporated,into,improve,training,method,leading,secondary,Simulation,experiments,were,carried,out,Gym,Carla,platform,verify,compare,models,results,showed,able,easily,states,which,could,traffic,safety,efficiency,above,reinforcement,significant,enhancing,vehicles
AB值:
0.54226
相似文献
机标中图分类号,由域田数据科技根据网络公开资料自动分析生成,仅供学习研究参考。