Combining Planning and Deep Reinforcement Learning in Tactical Decision Making for Autonomous Driving
Artikel i vetenskaplig tidskrift, 2019

Tactical decision making for autonomous driving is challenging due to the diversity of environments, the uncertainty in the sensor information, and the complex interaction with other road users. This article introduces a general framework for tactical decision making, which combines the concepts of planning and learning, in the form of Monte Carlo tree search and deep reinforcement learning. The method is based on the AlphaGo Zero algorithm, which is extended to a domain with a continuous state space where self-play cannot be used. The framework is applied to two different highway driving cases in a simulated environment and it is shown to perform better than a commonly used baseline method. The strength of combining planning and learning is also illustrated by a comparison to using the Monte Carlo tree search or the neural network policy separately.

Monte Carlo tree search

Autonomous driving

reinforcement learning

tactical decision making

Författare

Carl-Johan E Hoel

Chalmers, Mekanik och maritima vetenskaper, Fordonsteknik och autonoma system

Volvo Group

Katherine Driggs-Campbell

University of Illinois

Krister Wolff

Chalmers, Mekanik och maritima vetenskaper, Fordonsteknik och autonoma system

Leo Laine

Volvo Group

Chalmers, Mekanik och maritima vetenskaper

Mykel J. Kochenderfer

Stanford University

IEEE Transactions on Intelligent Vehicles

23798858 (eISSN)

Vol. 5 2 294-305

Ämneskategorier

Annan data- och informationsvetenskap

Systemvetenskap

Datavetenskap (datalogi)

Styrkeområden

Transport

DOI

10.1109/TIV.2019.2955905

Mer information

Senast uppdaterat

2020-11-27