|
A Novel Steering Control for Real Autonomous Vehicles via PI Adaptive
" B' \) P2 c2 w& N `Dynamic Programming / o/ h! b2 a) d9 O
! \8 u4 q* \5 e
6 H/ Q( \0 V! Y% I. z
$ v4 X: ^8 w1 K) C" |
A novel steering controller for autonomous vehicles is proposed in this paper. For autonomous vehicles, the efficiency
. J7 m2 X7 q# Pis an important indicator, in order to reduce the training cost and improve the real-time performance while the car is running on
4 ~6 h# c% ~/ i8 cthe road, Offline plus online training is proposed in this paper. Offline plus online training is presented to build a model network ; j: X8 C9 u P& v
of vehicle steering and correct it in real time. Then we adopt a policy iteration adaptive dynamic programming (ADP) to get the. Q! W G* q7 V- H2 y( B- F$ }7 W
optimal control law and deploy it to autonomous vehicles. Eventually, the performance of the controller was evaluated using " q, a6 ?9 D8 f2 d& u+ q
Matlab simulation. Results verify the feasibility of the theory and show the steering controller is one kind of effective, smooth
& `' S8 }0 H: l' {controller for autonomous vehicles' system. 6 {$ `8 `: G, L2 ^. v
9 P# o- I! Y( {; T# ]
& t+ [4 V( B. }$ T3 y
|