A Novel Steering Control for Real Autonomous Vehicles via PI Adaptive
- `: X7 Y1 j; A- ~7 _; D5 L' O% NDynamic Programming
* t9 D6 k* ]; V# J- i$ T o8 `" K8 L3 L/ ] W& \* q
* \% O2 F3 a* z
- G( D! z5 N$ s5 {6 _# Y, h" P' P$ \0 @+ a' c9 b
/ }* ^0 S" X5 n0 A. K4 d/ Z- j, Y' LA novel steering controller for autonomous vehicles is proposed in this paper. For autonomous vehicles, the efficiency - | p2 N, p% B
is an important indicator, in order to reduce the training cost and improve the real-time performance while the car is running on
2 {/ m8 K7 N3 s: `the road, Offline plus online training is proposed in this paper. Offline plus online training is presented to build a model network 8 }+ H7 ~) W; i7 \2 v( L2 H( o
of vehicle steering and correct it in real time. Then we adopt a policy iteration adaptive dynamic programming (ADP) to get the
7 N5 j. m- f: [optimal control law and deploy it to autonomous vehicles. Eventually, the performance of the controller was evaluated using
) @% @% j6 N+ R" H5 \+ U) NMatlab simulation. Results verify the feasibility of the theory and show the steering controller is one kind of effective, smooth 6 L9 P+ o' L& R6 h6 R% d
controller for autonomous vehicles' system. ( D/ B! v6 O" j9 C
{/ P: w. m; r0 p3 _. \9 L2 k
6 T- I W6 n: `6 k; A' @
|