Optimization of Global Chassis Control Variables

The paper presents a global chassis control (GCC) optimization approach using a gradient-based optimal control algorithm. The goal is to find optimal actions of various actuators such as active steering and active differential, which ensure satisfying the optimization criterion (e.g. trajectory following error minimization) subject to different equality and inequality constraints on state and control variables. The optimization algorithm is based on an exact gradient method, where the cost function gradient is calculated by using a backpropagation-through-time-like algorithm. The proposed GCC optimization approach is illustrated on an example of double lane change maneuver using rear active steering and/or rear active differential actuators.