A new optimization framework for robot motion planning

It isn't easy for a robot to find its way out of a maze. Picture these machines trying to traverse a kid's playroom to reach the kitchen, with miscellaneous toys scattered across the floor and furniture blocking some potential paths. This messy labyrinth requires the robot to calculate the most optimal journey to its destination, without crashing into any obstacles. What is the bot to do?