CaltechAUTHORS
  A Caltech Library Service

Voluntary lane-change policy synthesis with reactive control improvisation

Ge, Jin I. and Murray, Richard M. (2018) Voluntary lane-change policy synthesis with reactive control improvisation. . (Unpublished) https://resolver.caltech.edu/CaltechAUTHORS:20191029-152953038

[img] PDF - Submitted Version
See Usage Policy.

4Mb

Use this Persistent URL to link to this item: https://resolver.caltech.edu/CaltechAUTHORS:20191029-152953038

Abstract

In this paper, we propose reactive control improvisation to synthesize voluntary lane-change policy that meets human preferences under given traffic environments. We first train Markov models to describe traffic patterns and the motion of vehicles responding to such patterns using traffic data. The trained parameters are calibrated using control improvisation to ensure the traffic scenario assumptions are satisfied. Based on the traffic pattern, vehicle response models, and Bayesian switching rules, the lane-change environment for an automated vehicle is modeled as a Markov decision process. Based on human lane-change behaviors, we train a voluntary lane-change policy using explicit-duration Markov decision process. Parameters in the lane-change policy are calibrated through reactive control improvisation to allow an automated car to pursue faster speed while maintaining desired frequency of lane-change maneuvers in various traffic environments.


Item Type:Report or Paper (Discussion Paper)
Related URLs:
URLURL TypeDescription
https://www.cds.caltech.edu/~murray/preprints/gm18-cdc_s.pdfAuthorArticle
ORCID:
AuthorORCID
Ge, Jin I.0000-0001-6429-9337
Murray, Richard M.0000-0002-5785-7481
Additional Information:To appear, 2018 Conference on Decision and Control (CDC). This work is supported by NSF VeHiCal project 1545126. The authors would like to thank the support from NSF CPS Frontier project 1545126, and thank the insightful discussions with Prof. Sanjit Seshia and Prof. Gabor Orosz.
Funders:
Funding AgencyGrant Number
NSFCNS-1545126
Record Number:CaltechAUTHORS:20191029-152953038
Persistent URL:https://resolver.caltech.edu/CaltechAUTHORS:20191029-152953038
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:99544
Collection:CaltechAUTHORS
Deposited By: Tony Diaz
Deposited On:29 Oct 2019 22:41
Last Modified:29 Oct 2019 22:41

Repository Staff Only: item control page