Type
Conference PaperAuthors
Li, GuohaoMüller, Matthias

Casser, Vincent Michael
Smith, Neil
Michels, Dominik L.
Ghanem, Bernard

KAUST Department
Electrical Engineering ProgramVisual Computing Center (VCC)
Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
Computer Science Program
Date
2019-06Preprint Posting Date
2018-03-03Permanent link to this record
http://hdl.handle.net/10754/627342
Metadata
Show full item recordAbstract
Recent work has explored the problem of autonomous navigation by imitating a teacher and learning an end-to-end policy, which directly predicts controls from raw images. However, these approaches tend to be sensitive to mistakes by the teacher and do not scale well to other environments or vehicles. To this end, we propose Observational Imitation Learning (OIL), a novel imitation learning variant that supports online training and automatic selection of optimal behavior by observing multiple imperfect teachers. We apply our proposed methodology to the challenging problems of autonomous driving and UAV racing. For both tasks, we utilize the Sim4CV simulator that enables the generation of large amounts of synthetic training data and also allows for online learning and evaluation. We train a perception network to predict waypoints from raw image data and use OIL to train another network to predict controls from these waypoints. Extensive experiments demonstrate that our trained network outperforms its teachers, conventional imitation learning (IL) and reinforcement learning (RL) baselines and even humans in simulation.Citation
Li, G., Mueller, M., Michael Casser, V., Smith, N., Michels, D., & Ghanem, B. (2019). OIL: Observational Imitation Learning. Robotics: Science and Systems XV. doi:10.15607/rss.2019.xv.005Sponsors
This work was supported by the King Abdullah University of Science and Technology (KAUST) Office of Sponsored Research through the Visual Computing Center (VCC) funding.Publisher
arXivConference/Event name
Robotics: Science and Systems XVarXiv
1803.01129Additional Links
http://www.roboticsproceedings.org/rss15/p05.htmlRelations
Is Part Of:- [Project]
OIL: Observational Imitation Learning. URL: https://sites.google.com/kaust.edu.sa/oil/
- [Video]
OIL: Observational Imitation Learning. URL: https://youtu.be/_rhq8a0qgeg
Embedded External Content
ae974a485f413a2113503eed53cd6c53
10.15607/RSS.2019.XV.005