Weight-Free Multi-Objective Predictive Cruise Control of Autonomous Vehicles in Integrated Perturbation Analysis and Sequential Quadratic Programming Optimization Framework

Adaptive cruise control of autonomous vehicles can be posed as a multi-objective optimization problem where several conflicting criteria, e.g., fuel economy, tracking capability, ride comfort, and safety, need to be satisfied simultaneously. In order to reconcile these conflicting criteria, this paper presents a novel multi-objective predictive cruise control (MOPCC) approach in the feasible perturbation-based real-time iterative optimization framework. The longitudinal dynamics of vehicles are described as nonlinear car-tracking models. The new cost function for MOPCC is defined as the distance of the criteria vector to the vector of separately minimized criteria (i.e., a utopia point of the criteria). The weight-free MOPCC is then obtained by solving a constrained nonlinear optimal control problem in receding horizon fashion. Due to the difficulty in solving the optimization problem, the integrated perturbation analysis and sequential quadratic programming (InPA-SQP) is employed to compute the cruise controller. The merit of the proposed MOPCC is that it can systematically handle different cruise scenarios regardless of the weights of the predictive cruise control (PCC) criteria. Several driving cases are used to demonstrate the effectiveness and benefits of the proposed approach via comparing to weighted PCC approaches.