TY - GEN
T1 - End-to-End Drive By-Wire PID Lateral Control of an Autonomous Vehicle
AU - Baskaran, Akash
AU - Talebpour, Alireza
AU - Bhattacharyya, Shankar
PY - 2020
Y1 - 2020
N2 - The number of autonomous vehicles with advanced driver assistance systems have been increasing multi-fold. These technologies have reduced the work of the driver and have increased the safety of roads. Though a lot work has been done on development of autonomous vehicles, not much attention has been given to the millions of existing cars without these features. In this paper, we propose a method to implement level 2 autonomy in vehicles without Advanced Driver assistance systems. In this work, steering control of vehicles using voltage spoofing (can be extended to throttle and braking modules), development of PID controllers for the modules, and implementation of end-to-end driving to enable autonomous applications have been discussed. By searching for the stabilizing set to find the controller parameters Kp, Ki, and Kd, the system response has been improved and by implementing transfer learning, training data required has been reduced, and thus end-to-end driving with comparable results have been obtained.
AB - The number of autonomous vehicles with advanced driver assistance systems have been increasing multi-fold. These technologies have reduced the work of the driver and have increased the safety of roads. Though a lot work has been done on development of autonomous vehicles, not much attention has been given to the millions of existing cars without these features. In this paper, we propose a method to implement level 2 autonomy in vehicles without Advanced Driver assistance systems. In this work, steering control of vehicles using voltage spoofing (can be extended to throttle and braking modules), development of PID controllers for the modules, and implementation of end-to-end driving to enable autonomous applications have been discussed. By searching for the stabilizing set to find the controller parameters Kp, Ki, and Kd, the system response has been improved and by implementing transfer learning, training data required has been reduced, and thus end-to-end driving with comparable results have been obtained.
KW - End-to-end drive
KW - PID controller
KW - Transfer learning
UR - http://www.scopus.com/inward/record.url?scp=85075641754&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85075641754&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-32520-6_29
DO - 10.1007/978-3-030-32520-6_29
M3 - Conference contribution
AN - SCOPUS:85075641754
SN - 9783030325190
T3 - Advances in Intelligent Systems and Computing
SP - 365
EP - 376
BT - Proceedings of the Future Technologies Conference, FTC 2019 Volume 1
A2 - Arai, Kohei
A2 - Bhatia, Rahul
A2 - Kapoor, Supriya
PB - Springer
T2 - 4th Future Technologies Conference, FTC 2019
Y2 - 24 October 2019 through 25 October 2019
ER -