A unified approach to rollover prevention based on control allocation
This paper presents a novel technique for transforming driver commands into chassis control signals in the context of rollover prevention when a vehicle with high mass centre is driven at high speed on a road of increasing curvature. The problem is complicated by the need to address several issues: un-tripped and tripped rollover, the trade-off between path following and rollover prevention, and also need for a seamless transition between keeping the driver in the control-loop and allowing an electronic safety system to make a fully autonomous intervention. Different options for control moderation and actuator control allocation (CA) are investigated. The aim is to establish best-case methods for autonomous and driver-adaptive interventions based on different road conditions. From extended simulation test, it is proposed that two available CA approaches can be used synergeticly and efficiently to solve this problem.