|
A Novel Steering Control for Real Autonomous Vehicles via PI Adaptive # Y! N9 U( w4 ?$ f6 W' A- S
Dynamic Programming
2 L$ N1 M8 v' _% }' _( v. f9 i& d( p, e
1 @! E1 O5 w1 r% {
& H; D ^5 t8 p2 V4 xA novel steering controller for autonomous vehicles is proposed in this paper. For autonomous vehicles, the efficiency # F* Z: C. ~+ l8 \. P* u) C) ]$ Q
is an important indicator, in order to reduce the training cost and improve the real-time performance while the car is running on
5 G/ p0 P/ I! G9 h7 o4 G7 q& bthe road, Offline plus online training is proposed in this paper. Offline plus online training is presented to build a model network * n# S& V t# f
of vehicle steering and correct it in real time. Then we adopt a policy iteration adaptive dynamic programming (ADP) to get the8 x" }/ O# y; m# h3 `, r: ~
optimal control law and deploy it to autonomous vehicles. Eventually, the performance of the controller was evaluated using
$ B- F0 l0 y+ {, O5 n, t# _Matlab simulation. Results verify the feasibility of the theory and show the steering controller is one kind of effective, smooth
, e% C: j6 G) m+ D, ?0 P# Icontroller for autonomous vehicles' system. 9 n# Q! p) M [- V. Z2 G V* ^
0 V/ _) L( M4 S5 T
' l+ L+ d7 Z! g4 X& b |