|
A Novel Steering Control for Real Autonomous Vehicles via PI Adaptive ) U' D E: \; r1 Q- J
Dynamic Programming 5 C, D5 A E" q( r
1 v# I/ D5 {" F9 e2 z& V
% [. ^: E* y) n" }) w- N1 x
4 v# D* X# [% N; F6 B
4 Q& O. x% V' G4 H0 A" b5 Y9 y3 F e8 R$ ~, a) C }; q9 B
A novel steering controller for autonomous vehicles is proposed in this paper. For autonomous vehicles, the efficiency
3 v! }6 D% m( y! u6 [7 l" qis an important indicator, in order to reduce the training cost and improve the real-time performance while the car is running on
" Z* E! p$ J! C5 U" Bthe road, Offline plus online training is proposed in this paper. Offline plus online training is presented to build a model network / k% J0 x6 u' B6 f* I4 |
of vehicle steering and correct it in real time. Then we adopt a policy iteration adaptive dynamic programming (ADP) to get the
f, P6 {- V" l4 U i& z# Q+ Roptimal control law and deploy it to autonomous vehicles. Eventually, the performance of the controller was evaluated using
* [3 b5 _9 b. v% _2 K. Q/ s; H2 F3 bMatlab simulation. Results verify the feasibility of the theory and show the steering controller is one kind of effective, smooth
8 U4 Z2 O: a; G2 E- `- ]controller for autonomous vehicles' system.
% a2 B5 U! ?; u4 f8 {6 A$ G9 m2 `8 I3 y9 q. {8 d6 n# U& V
/ Z# ]! @7 ?2 v: T2 v
|