Dynamic Model and Control Strategies of Autonomous Forklift: a Review

Autonomous forklifts have become important parts of smart warehouse systems driven by Industry 4.0, with market projections reaching $1.52 billion by 2032. This is partly due to the high accident rate of conventional forklifts caused by rollovers and tip overs, which reaches 25.3%. This systematic review examines dynamic modeling and control strategies for autonomous forklifts, focusing on roll stabilization and motion control. The review identifies critical research gaps in current autonomous forklift technology. Roll stabilization research employs various evaluation indices including Lateral Transfer Ratio (LTR), energy-based indices, and phase plane methods, with dynamic models ranging from 3-DOF to 7-DOF configurations. Model Predictive Control (MPC) emerges as the dominant control strategy, appearing in multiple studies due to its ability to handle constraints and multi-objective optimization. Motion control approaches include MPC-based methods, Active Disturbance Rejection Control (ADRC), and specialized velocity control, addressing challenges such as nonholonomic constraints and trajectory tracking accuracy. However, a significant research gap exists: current studies treat roll stabilization, tip-over stabilization and motion control as separate problems, potentially generating conflicting control objectives and unsafe operations during aggressive maneuvers. This review emphasizes the critical need for integrated control frameworks that simultaneously optimize two or more of: trajectory tracking, roll stability, energy efficiency, and task completion time, representing a transformative opportunity for advancing autonomous forklift technology toward safer and more efficient industrial applications.