Near time-optimal control of racing vehicles
1989; Elsevier BV; Volume: 25; Issue: 6 Linguagem: Inglês
10.1016/0005-1098(89)90052-6
ISSN1873-2836
AutoresDavid Metz, David M. Williams,
Tópico(s)Electric and Hybrid Vehicle Technologies
ResumoThe racing driver constantly tries to select control inputs (steering, braking and acceleration) so as to minimize lap times. The car itself is governed by non-linear differential equations of motion which are geometrically (via the track boundaries) and mechanically (via engine power, braking and tire limits) constrained. The driver selects control inputs which produce minimum lap times by operating the vehicle as near as possible to its limit of control. In the present paper, a realistic series of tire, vehicle handling and engine/braking models is developed for open wheeled, single seat racing cars of the Indy car/Formula One variety. The model is stimulated over an actual section of the Formula One Circuit Paul Ricard in southern France, and a quasioptimization routine is developed to compute the controls required for minimization of time through this course section. Results are compared with actual performance data measured from a Formula One car driven by three expert racing drivers (Andretti, Peterson and Stewart). Results showed good qualitative agreement between measured and computed performance. An examination of driver performance and control selection shows several high-order, fine control inputs which are present in both experimental and theoretical results, and allows development of explanations for driver actions which often appear counterintuitive from the racing engineer's point of view.
Referência(s)