MILP optimal path planning for real-time applications

Abstract

This paper presents several efficient solution techniques specific to the optimal path planning of an autonomous vehicle. Mixed-integer linear programming (MILP) is the underlying problem formulation, from which an optimal solution can be obtained through the use of a commercially available MILP solver such as CPLEX. The solution obtained is optimal in terms of the cost function specified in terms of fuel, time, altitude, etc. Several techniques are introduced to reduce the complexity of the underlying mathematical problems as to help make the path planning approach suitable for running in a mission-critical real-time environment. Some of these techniques may be applicable to other optimal path planning approaches