|
A Novel Steering Control for Real Autonomous Vehicles via PI Adaptive 7 l1 S. e5 y+ S
Dynamic Programming & o2 A. h8 I& V
' Q( ~ ^# l+ f Y$ X2 Y9 P* O: \) G0 ^
+ h# C8 ^2 P) _+ E+ g* _) Y! ^9 h
A novel steering controller for autonomous vehicles is proposed in this paper. For autonomous vehicles, the efficiency
" W' k$ V: Y7 u: ~is an important indicator, in order to reduce the training cost and improve the real-time performance while the car is running on . P6 ~1 {. x" D$ Q/ B$ z
the road, Offline plus online training is proposed in this paper. Offline plus online training is presented to build a model network
% u7 \, Z/ R0 D& p% O8 Y6 S) {of vehicle steering and correct it in real time. Then we adopt a policy iteration adaptive dynamic programming (ADP) to get the( v0 o4 N8 w& @
optimal control law and deploy it to autonomous vehicles. Eventually, the performance of the controller was evaluated using 5 z9 b5 @4 D) d% o- x @* V' a& I
Matlab simulation. Results verify the feasibility of the theory and show the steering controller is one kind of effective, smooth # h( ^3 `. j3 S. A" i% a" H
controller for autonomous vehicles' system. 4 v, ?" a. i! z Z- ~& p% W8 O5 y
: S9 E7 x. G6 P0 j, h# z
7 J. u$ c, w; K; e7 I: H. K5 w |