Learning the Problem-Optimum Map: Analysis and Application to Global Optimization in Robotics
This work addresses the need for fast, near-optimal solutions in robotics applications, such as real-time motion planning, by providing a practical framework that significantly reduces computation time compared to traditional methods.
The paper tackles the challenge of real-time global optimization in robotics by developing a data-driven framework that precomputes solutions to sample problems and adapts them online, achieving near-globally optimal solutions orders of magnitude faster than standard methods, with sub-millisecond computation times for collision-free inverse kinematics on a standard PC.
This paper describes a data-driven framework for approximate global optimization in which precomputed solutions to a sample of problems are retrieved and adapted during online use to solve novel problems. This approach has promise for real-time applications in robotics, since it can produce near-globally optimal solutions orders of magnitude faster than standard methods. This paper establishes theoretical conditions on how many and where samples are needed over the space of problems to achieve a given approximation quality. The framework is applied to solve globally optimal collision-free inverse kinematics (IK) problems, wherein large solution databases are used to produce near-optimal solutions in sub-millisecond time on a standard PC.