Abstract
Keywords: Cognitive Automobile, Safety Concept, Vehicle Motion Control, Knowledge-based Diagnosis, DGPS-based Path Planning
Autonomous driving inclusive accelerating, braking, steering and other advanced maneuvers, such as lane change and platooning should be exactly executed in the cognitive automobile. Such object is required to be accomplished by the vehicle actuators precisely, in limited delay and without errors by means of variable control algorithms.
Because this control process completely takes the vehicle motion in hand and brings the fatal error to people and the vehicle in case of a false implementation, it should be secured. On the other hand a fallback solution is needed in case of emergency to bring the vehicle back to the safety status such as 'stop' if the driver cannot handle the situation in right time or manner.
This article discusses how to achieve the reliable vehicle motion control and ensure its safety processing. The redundant system design is the basis for the later mentioned capability based vehicle control strategy and the safety concept. Before the trajectory commands are generated from the cognition units, the actual vehicle's physical boundary has to be taken into consideration. The reliable execution of the drive task is based on whether it is under the vehicle's actual capability, since otherwise it will be certainly not or not correctly carried out. In this step the precise prediction of the vehicle ability and dynamic potential would play the important roll. Following the restricting or adjusting of the driving task it will be deconstructed into elementary maneuvers, which will be arranged and executed subsequently and accordingly on the separate actuators. Knowledge-based diagnosis will be applied to observe the status of the hardware and its degree of performance. In case of failure in the sensor-controller-actuator loop, it should indicate the failure location and switch the controller to its backup solution. In the case of 'blind' mode, which means the loss of the signal from the vision- and cognition-unit, the task is to try to keep the vehicle in the right path during the fail-safe deceleration. For this purpose a path planning algorithm is developed, which is updated regularly "online" according to the position and therefore guarantees that the vehicle knows the track for the next few seconds. This is realized with the use of a digital map database and a differential GPS receiver.