Sentio: Driver-in-the-Loop Forward Collision Warning Using Multisample Reinforcement Learning
KAUST Grant NumberSensor Innovation research program
Permanent link to this recordhttp://hdl.handle.net/10754/668426
MetadataShow full item record
AbstractThanks to the adoption of more sensors in the automotive industry, context-aware Advanced Driver Assistance Systems (ADAS) become possible. On one side, a common thread in ADAS applications is to focus entirely on the context of the vehicle and its surrounding vehicles leaving the human (driver) context out of consideration. On the other side, and due to the increasing sensing capabilities in mobile phones and wearable technologies, monitoring complex human context becomes feasible which paves the way to develop driver-in-the-loop context-aware ADAS that provide personalized driving experience. In this paper, we propose Sentio 1 ; a Reinforcement Learning based algorithm to enhance the Forward Collision Warning (FCW) system leading to Driver-in-the-Loop FCW system. Since the human driving preference is unknown a priori, varies between different drivers, and moreover, varies across time for the same driver, the proposed Sentio algorithm needs to take into account all these variabilities which are not handled by the standard reinforcement learning algorithms. We verified the proposed algorithm against several human drivers. Our evaluation, across distracted human drivers, shows a significant enhancement in driver experience—compared to standard FCW systems—reflected by an increase in the driver safety by 94.28%, an improvement in the driving experience by 20.97%, a decrease in the false negatives from 55.90% down to 3.26%, while adding less than 130 ms runtime execution overhead.
CitationElmalaki, S., Tsai, H.-R., & Srivastava, M. (2018). Sentio. Proceedings of the 16th ACM Conference on Embedded Networked Sensor Systems. doi:10.1145/3274783.3274843
SponsorsThis research was supported in part by the NIH Center of Excellence for Mobile Sensor Data-to-Knowledge (MD2K) under award 1-U54EB020404-01, the U.S. Army Research Laboratory and the UK Ministry of Defense under Agreement Number W911NF-16-3-0001, the National Science Foundation under awards #OAC-1640813 and CNS-1329755, and the King Abdullah University of Science and Technology (KAUST) through its Sensor Innovation research program. The Microsoft Research PhD Fellowship has supported Salma Elmalaki. Any findings in this material are those of the author(s) and do not reflect the views of any of the above funding agencies. The U.S. and U.K. Governments are authorized to reproduce and distribute reprints for Government purposes notwithstanding any copyright notation hereon.
Conference/Event name16th ACM Conference on Embedded Networked Sensor Systems, SENSYS 2018