|
A Novel Steering Control for Real Autonomous Vehicles via PI Adaptive ' ]5 E9 _6 ~' G/ C% @* A' ]0 x2 o# y: {
Dynamic Programming
3 w5 s* H) ]4 b, R; G8 @+ n6 M
# h. K9 u0 w. _6 N: v4 }% w* K, u5 J& Y
9 {3 h9 Z4 v ` [) g0 o& b- {A novel steering controller for autonomous vehicles is proposed in this paper. For autonomous vehicles, the efficiency ^/ v! c' o7 e( C$ C: U2 a
is an important indicator, in order to reduce the training cost and improve the real-time performance while the car is running on
) L. C3 k2 G* mthe road, Offline plus online training is proposed in this paper. Offline plus online training is presented to build a model network . Q/ T# w. S/ C; _6 p
of vehicle steering and correct it in real time. Then we adopt a policy iteration adaptive dynamic programming (ADP) to get the4 W* W. G* b% e: j) S7 o- C$ H
optimal control law and deploy it to autonomous vehicles. Eventually, the performance of the controller was evaluated using
7 O+ m4 g/ V; Y5 `- s4 J+ QMatlab simulation. Results verify the feasibility of the theory and show the steering controller is one kind of effective, smooth
# l& b# a# R0 K& ?* Tcontroller for autonomous vehicles' system. 5 \' C. O, M. ?4 f
& c* O4 O$ H* w; R) ?
' F8 x% s( q% t, N! c6 T" ~ |