Reference Tracking Optimization With Obstacle Avoidance via Task Prioritization for Automated Driving

Obstacle avoidance is a fundamental operation for automated driving and its formulation traditionally originates from robotics decision making control fields. Given the high complexity required to compute an obstacle-free trajectory, this usually demanded lower frequency planning layer that provides then trajectory reference be followed by higher layer. As result, whenever replanning needed (for example, due new detected obstacle), must wait planned generated. In paper, we propose novel methodology approach obstacle already in layer, which allows prompter response. particular, show how tracking can integrated, thus with no need switch among different controllers, based on null-space behavioral approach, implemented (possibly nonlinear) model predictive scheme. We demonstrate practical implementation of proposed employing two vehicle dynamic models four (urban highway) scenarios. Furthermore, provide sensitivity analysis understand parameters choice affects behavior.