A Game-Theoretic Reinforcement Learning Approach for Adaptive Interaction at Intersections
Authors:
- 6 November 2020
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 4451-4456
Abstract
In this paper, we propose a bi-level algorithm for motion planning at intersections based on a scheme of reasoning game theory and heuristic reinforcement learning. In the upper level, a recurrent neural network is introduced to estimate the type of opponent agent. In the lower level, Q-networks are selectively connected to implement the game with different type. Then the ego agent could update its estimation step-by-step and conclude correspond action from historical joint state. The simulation results show that the bi-level controller improves pass times and collision avoidance performance.Keywords
Funding Information
- National Natural Science Foundation of China
This publication has 17 references indexed in Scilit:
- Decentralized non-communicating multiagent collision avoidance with deep reinforcement learning2017
- A Survey of Motion Planning and Control Techniques for Self-Driving Urban VehiclesIEEE Transactions on Intelligent Vehicles, 2016
- An Intersection Game-Theory-Based Traffic Control Algorithm in a Connected Vehicle Environment2015
- Human-level control through deep reinforcement learningNature, 2015
- A review of game-theoretic models of road user behaviourAccident Analysis & Prevention, 2014
- A Dynamic Level-k Model in Sequential GamesManagement Science, 2013
- Estimation of Multivehicle Dynamics by Considering Contextual InformationIEEE Transactions on Robotics, 2012
- Comparing Models of Strategic Thinking in Van Huyck, Battalio, and Beil's Coordination GamesJournal of the European Economic Association, 2009
- Cognition and Behavior in Two-Person Guessing Games: An Experimental StudyAmerican Economic Review, 2006
- Long Short-Term MemoryNeural Computation, 1997