Dynamically balanced trajectory generation for non-holonomic forklift vehicles
Leng Vongchanh, Alireza M. Mohammadi, Allison Kealy, Denny Oetomo · International Conference on Robotics and Automation · 2014
This paper proposes a trajectory planner for a non-holonomic wheeled mobile manipulator, specifically designed for a forklift vehicle (FLV). The planner addresses the high risk toppling, in both the lateral and sagittal directions, by integrating dynamic balance of the system in the design of the trajectory. The proposed trajectory extends the Dubins' and simple continuous curvature (SCC) paths, using time dependent parameters to build the path, while the dynamic balance is considered through the use of the zero moment point (ZMP) and a barycentric coordinate system to develop a scalar measure of toppling propensity. This geometric form was chosen to reduce the problem search space in consideration for industrial application. An computational illustration will compare existing paths and highlight the benefits for use with an FLV.