|
A Novel Steering Control for Real Autonomous Vehicles via PI Adaptive 7 Y+ I" ~, e g7 C
Dynamic Programming
; u& _: r3 ^ R- F( ^. ^5 Y" o* `1 N- @$ R3 @: G9 m m; G0 g/ `7 p
; n) F1 H# \) I" r1 J2 H+ b |5 P4 ]3 N/ T
A novel steering controller for autonomous vehicles is proposed in this paper. For autonomous vehicles, the efficiency 9 e* t! P$ D4 j1 f% @, @
is an important indicator, in order to reduce the training cost and improve the real-time performance while the car is running on ' F4 y; y6 K" T0 j. X3 P
the road, Offline plus online training is proposed in this paper. Offline plus online training is presented to build a model network 5 _' J( G3 Q1 w: V u6 F
of vehicle steering and correct it in real time. Then we adopt a policy iteration adaptive dynamic programming (ADP) to get the/ H% s9 V9 G0 |$ c2 h
optimal control law and deploy it to autonomous vehicles. Eventually, the performance of the controller was evaluated using
* f3 A/ i/ W6 x7 I! B( nMatlab simulation. Results verify the feasibility of the theory and show the steering controller is one kind of effective, smooth
) j# _8 s& v2 O! M4 l1 J3 mcontroller for autonomous vehicles' system. 1 F' g* P" g& Q9 k9 w: _
. b) o! B* m! l" S+ q' P* `2 e4 ]% o5 a3 z% _
|