|
A Novel Steering Control for Real Autonomous Vehicles via PI Adaptive - C2 j) I0 ~/ c5 I3 l
Dynamic Programming
2 F4 n# D: L3 I0 n# X5 K% w' s" F7 m; a6 c7 C! {& n
* P& S. B2 d5 a) P! k
; d6 X6 H2 _; D: P- w; fA novel steering controller for autonomous vehicles is proposed in this paper. For autonomous vehicles, the efficiency 2 _ `& ?0 h& Y/ M9 q& R% t1 M
is an important indicator, in order to reduce the training cost and improve the real-time performance while the car is running on
8 q5 |% L2 a- j$ `- rthe road, Offline plus online training is proposed in this paper. Offline plus online training is presented to build a model network 1 u1 e. t; {4 S/ x) c, C6 u& I/ k; E8 I/ Q
of vehicle steering and correct it in real time. Then we adopt a policy iteration adaptive dynamic programming (ADP) to get the
; Z4 `, c/ A: O/ B8 Eoptimal control law and deploy it to autonomous vehicles. Eventually, the performance of the controller was evaluated using ! G+ l9 m( y" v& t) F
Matlab simulation. Results verify the feasibility of the theory and show the steering controller is one kind of effective, smooth + h( n" K! e6 p' a$ v5 N$ N3 L; ~
controller for autonomous vehicles' system. ( p( e; M, N0 L V1 u r
* D9 q/ p. f+ \& d+ J3 n1 I
& h4 W/ g& p, j+ V1 Q |