TY - GEN
T1 - Self Training Autonomous Driving Agent
AU - Kotyan, Shashank
AU - Vargas, Danilo Vasconcellos
AU - Venkanna, U.
PY - 2019/9
Y1 - 2019/9
N2 - Intrinsically, driving is a Markov Decision Process which suits well the reinforcement learning paradigm. In this paper, we propose a novel agent which learns to drive a vehicle without any human assistance. We use the concept of reinforcement learning and evolutionary strategies to train our agent in a 2D simulation environment. Our model's architecture goes beyond the World Model's by introducing difference images in the autoencoder. This novel involvement of difference images in the auto-encoder gives a better representation of the latent space concerning the motion of the vehicle and helps an autonomous agent to learn more efficiently how to drive a vehicle. Results show that our method requires fewer (96% less) total agents, (87.5% less) agents per generations, (70% less) generations and (90% less) rollouts than the original architecture while achieving the same accuracy of the original.
AB - Intrinsically, driving is a Markov Decision Process which suits well the reinforcement learning paradigm. In this paper, we propose a novel agent which learns to drive a vehicle without any human assistance. We use the concept of reinforcement learning and evolutionary strategies to train our agent in a 2D simulation environment. Our model's architecture goes beyond the World Model's by introducing difference images in the autoencoder. This novel involvement of difference images in the auto-encoder gives a better representation of the latent space concerning the motion of the vehicle and helps an autonomous agent to learn more efficiently how to drive a vehicle. Results show that our method requires fewer (96% less) total agents, (87.5% less) agents per generations, (70% less) generations and (90% less) rollouts than the original architecture while achieving the same accuracy of the original.
UR - http://www.scopus.com/inward/record.url?scp=85073874223&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85073874223&partnerID=8YFLogxK
U2 - 10.23919/SICE.2019.8859883
DO - 10.23919/SICE.2019.8859883
M3 - Conference contribution
T3 - 2019 58th Annual Conference of the Society of Instrument and Control Engineers of Japan, SICE 2019
SP - 1456
EP - 1461
BT - 2019 58th Annual Conference of the Society of Instrument and Control Engineers of Japan, SICE 2019
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 58th Annual Conference of the Society of Instrument and Control Engineers of Japan, SICE 2019
Y2 - 10 September 2019 through 13 September 2019
ER -