Development of a nonlinear predictive controller for mitigation of motion sickness in autonomous vehicles through multi-objective control of lateral and roll dynamics

This paper presents the design and evaluation of a nonlinear predictive controller for vehicle path following, specifically aimed at mitigating motion sickness (MS). The controller's cost function incorporates key vehicle motion components – lateral, roll, and yaw motions – to reduce occupant discomfort. By managing path-following and MS-related variables concurrently, the control law enhances ride smoothness. The controller is designed using a linear vehicle model that includes lateral and roll dynamics and is tested on an 11-degree-of-freedom nonlinear full-vehicle model. Performance is assessed using three metrics that evaluate motion smoothness: The Integral RMS Jerk criterion, which measures the rate of change of acceleration (jerk), the Cumulative Absolute Acceleration criterion, and the standard deviation of jerks. Computer simulations in MATLAB/Simulink, conducted on a double-lane-change manoeuvre at both low and high speeds, demonstrate that the proposed controller reduces MS-related motion metrics. • Nonlinear predictive controller minimizes lateral, roll, and yaw motions. • Balances path-following with motion sickness mitigation, improving passenger comfort. • Assesses controller with Integral RMS Jerk, Cumulative Absolute Acceleration, and jerk SD. • Significant reduction in lateral and roll accelerations during lane changes. • Short prediction horizon enables real-time use in autonomous vehicle systems.