Corpus ID: 215416017

Learning Mixed-Integer Convex Optimization Strategies for Robot Planning and Control

@article{Cauligi2020LearningMC,
  title={Learning Mixed-Integer Convex Optimization Strategies for Robot Planning and Control},
  author={Abhishek Cauligi and Preston Culbertson and Bartolomeo Stellato and Dimitris Bertsimas and Mac Schwager and Marco Pavone},
  journal={ArXiv},
  year={2020},
  volume={abs/2004.03736}
}
  • Abhishek Cauligi, Preston Culbertson, +3 authors Marco Pavone
  • Published 2020
  • Engineering, Computer Science
  • ArXiv
  • Mixed-integer convex programming (MICP) has seen significant algorithmic and hardware improvements with several orders of magnitude solve time speedups compared to 25 years ago. Despite these advances, MICP has been rarely applied to real-world robotic control because the solution times are still too slow for online applications. In this work, we extend the machine learning optimizer (MLOPT) framework to solve MICPs arising in robotics at very high speed. MLOPT encodes the combinatorial part of… CONTINUE READING

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 33 REFERENCES

    LVIS: Learning from Value Function Intervals for Contact-Aware Robot Controllers

    VIEW 3 EXCERPTS

    Reactive Planar Manipulation with Convex Hybrid MPC

    VIEW 2 EXCERPTS

    Approximate hybrid model predictive control for multi-contact push recovery in complex environments

    VIEW 3 EXCERPTS

    GuSTO: Guaranteed Sequential Trajectory optimization via Sequential Convex Programming

    VIEW 1 EXCERPT

    Safe and Near-Optimal Policy Learning for Model Predictive Control using Primal-Dual Neural Networks

    VIEW 1 EXCERPT

    Learning Trajectories for Real- Time Optimal Control of Quadrotors

    VIEW 1 EXCERPT