Cornering Trajectory Planning Avoiding Slip for Differential-Wheeled Mobile Robots

摘要

In this article, we propose an efficient algorithm for cornering trajectory planning avoiding slip based on dynamics for differential-wheeled mobile robots (DWMRs). In general, slip between tires and the ground increases odometry errors. Hence, to consider not only full dynamics, including actuators, but also slip avoidance, we formulate a complex planning problem, in which the following essential constraints are imposed: longitudinal and lateral force constraints to avoid slip and motor control input bounds for DWMR dynamics, including actuators. Given a single corner, the proposed trajectory is divided into three turning sections, which are composed of multiple intervals. In order to satisfy all constraints, the intervals are then split into several subintervals, which are with constant or time-varying control inputs based on the bang-bang principle or on force constraint bounds for optimal acceleration. We then find two local minimum trajectories by analyzing the time trend w.r.t. the turning radius for cornering. These are the low-velocity decrease trajectory and the shortest distance trajectory. The time-optimal solution is determined by the minimum of the local minima. Both simulations and experiments are conducted to verify the effectiveness of the proposed algorithm.