Mixed-Integer and Conditional Trajectory Planning for an Autonomous Mining Truck in Loading/Dumping Scenarios: A Global Optimization Approach

Trajectory planning for a heavy-duty mining truck near the loading/dumping sites of an open-pit mine is difficult. As opposed to trajectory small-sized passenger car in parking lot, involves complex factors vehicle kinematics and environment. These make concerned scheme mixed-integer nonlinear program (MINLP) incorporated with conditional constraints (denoted as C-MINLP). MINLP solvers can neither deal nor find global optima real time. Instead solving C-MINLP directly, we build from-coarse-to-fine framework so that coupled difficulties (the mixed integral variables, constraints, demand optimality) are divided conquered. At coarse search stage, global-optimality-enhanced hybrid A * algorithm proposed near-optimal kinematic optimality considered. The further polished at refinement wherein nominal simplified small-scale NLP. solution NLP optimized trajectory, which does not violate C-MINLP. This indicates conversion from efficient help high-quality trajectory.