|
A Novel Steering Control for Real Autonomous Vehicles via PI Adaptive " Q" C. \; \% a- C* S) r0 `
Dynamic Programming * }+ q" Y2 N: e$ U, u. h
2 ~' O& u% w" s4 m- {; H8 f: }9 ]$ @. j/ x3 u0 v& }1 X
* T3 I) |3 I6 r3 e. D
A novel steering controller for autonomous vehicles is proposed in this paper. For autonomous vehicles, the efficiency
& l$ d, _ L- D2 r/ jis an important indicator, in order to reduce the training cost and improve the real-time performance while the car is running on
3 | e& f; k8 r4 ethe road, Offline plus online training is proposed in this paper. Offline plus online training is presented to build a model network
' m% y5 G- T- I& [5 a- S& E1 j, uof vehicle steering and correct it in real time. Then we adopt a policy iteration adaptive dynamic programming (ADP) to get the
3 |" |9 M4 O( l5 K3 a" V3 E- _optimal control law and deploy it to autonomous vehicles. Eventually, the performance of the controller was evaluated using 2 x! y2 z, `+ h: d* ~% L4 T( i0 [
Matlab simulation. Results verify the feasibility of the theory and show the steering controller is one kind of effective, smooth ' I5 J4 ~& [2 V7 R$ b
controller for autonomous vehicles' system. 8 q! ~# R4 I! n" {
$ R' E, u. W7 F, K' X1 a
7 G3 S& Y# B n( P
| 
IP卡
狗仔卡
发表于 2020-11-10 15:34
转播
淘帖
分享
收藏
支持
反对
微信
提升卡
置顶卡
沉默卡
喧嚣卡
变色卡
抢沙发
显身卡
