CoCo: Online Mixed-Integer Control Via Supervised Learning

• Published in: IEEE robotics and automation letters Vol. 7; no. 2; pp. 1447 - 1454
• Main Authors: Cauligi, Abhishek, Culbertson, Preston, Schmerling, Edward, Schwager, Mac, Stellato, Bartolomeo, Pavone, Marco
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 01.04.2022; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Classifiers; Combinatorial analysis; Computational geometry; Convex functions; Convexity; data-driven optimization; Mixed integer; mixed-integer convex programming; Motion planning; Neural networks; Optimization; Optimization and optimal control; Parameters; Performance degradation; Planning; Robot control; Robot dynamics; Robotics; Robots; Strategy; Supervised learning; Task analysis; Trajectory optimization; Upper bound
• ISSN: 2377-3766; 2377-3766
• DOI: 10.1109/LRA.2021.3135931

Many robotics problems, from robot motion planning to object manipulation, can be modeled as mixed-integer convex program (MICPs). However, state-of-the-art algorithms are still unable to solve MICPs for control problems quickly enough for online use and existing heuristics can typically only find suboptimal solutions that might degrade robot performance. In this work, we turn to data-driven methods and present the Combinatorial Offline, Convex Online (CoCo) algorithm for quickly finding high quality solutions for MICPs. CoCo consists of a two-stage approach. In the offline phase, we train a neural network classifier that maps the problem parameters to a logical strategy , which we define as the discrete arguments and relaxed big-M constraints associated with the optimal solution for that problem. Online, the classifier is applied to select a candidate logical strategy given new problem parameters; applying this logical strategy allows us to solve the original MICP as a convex optimization problem. We show through numerical experiments how CoCo finds near optimal solutions to MICPs arising in robot planning and control with 1 to 2 orders of magnitude solution speedup compared to other data-driven approaches and solvers.