|
A Novel Steering Control for Real Autonomous Vehicles via PI Adaptive 8 p2 u! |' _. c4 R( i
Dynamic Programming * Z7 E' U: w/ \ o
. L U8 y8 R! w$ M
; Z7 U, q u" G, v) F4 J1 \ {/ s8 Y) } I+ j
6 P6 f& Y% w3 w; ^5 x
' t; E6 k' d% P- @A novel steering controller for autonomous vehicles is proposed in this paper. For autonomous vehicles, the efficiency
2 {; E% H0 M( a7 Ais an important indicator, in order to reduce the training cost and improve the real-time performance while the car is running on / ?: o7 n/ t6 w7 f7 ~! a
the road, Offline plus online training is proposed in this paper. Offline plus online training is presented to build a model network
4 ]5 T6 b% y6 w) `, F+ C/ ]of vehicle steering and correct it in real time. Then we adopt a policy iteration adaptive dynamic programming (ADP) to get the$ B4 y% o4 H$ J, w
optimal control law and deploy it to autonomous vehicles. Eventually, the performance of the controller was evaluated using
) Z- t( f3 g7 U/ Y { `Matlab simulation. Results verify the feasibility of the theory and show the steering controller is one kind of effective, smooth
+ k1 }# E3 l. m2 W1 O$ E9 Scontroller for autonomous vehicles' system. " g, g( Q" g4 g2 B: |2 \3 B
8 V, E% q N$ C
4 j: K1 x9 l% f4 y" r" l) k7 X |