Model-Based Control with Sparse Neural Dynamics
Preprint
Abstract
Learning predictive models from observations using deep neural networks (DNNs) is a promising new approach to many real-world planning and control problems. However, common DNNs are too unstructured for effective planning, and current control methods typically rely on extensive sampling or local gradient descent. In this paper, we propose a new framework for integrated model learning and predictive control that is amenable to efficient optimization algorithms. Specifically, we start with a ReLU neural model of the system dynamics and, with minimal losses in prediction accuracy, we gradually sparsify it by removing redundant neurons. This discrete sparsification process is approximated as a continuous problem, enabling an end-to-end optimization of both the model architecture and the weight parameters. The sparsified model is subsequently used by a mixed-integer predictive controller, which represents the neuron activations as binary variables and employs efficient branch-and-bound algorithms. Our framework is applicable to a wide variety of DNNs, from simple multilayer perceptrons to complex graph neural dynamics. It can efficiently handle tasks involving complicated contact dynamics, such as object pushing, compositional object sorting, and manipulation of deformable objects. Numerical and hardware experiments show that, despite the aggressive sparsification, our framework can deliver better closed-loop performance than existing state-of-the-art methods.Keywords
All Related Versions
This publication has 14 references indexed in Scilit:
- Grounding DINO: Marrying DINO with Grounded Pre-training for Open-Set Object Detection2024
- Shortest Paths in Graphs of Convex SetsSIAM Journal on Optimization, 2024
- Learning where to trust unreliable models in an unstructured world for deformable object manipulationScience Robotics, 2021
- The Surprising Effectiveness of Linear Models for Visual Foresight in Object Pile Manipulation2021
- Mastering Atari, Go, chess and shogi by planning with a learned modelNature, 2020
- Warm Start of Mixed-Integer Programs for Model Predictive Control of Hybrid SystemsIEEE Transactions on Automatic Control, 2020
- Structured Pruning of Deep Convolutional Neural NetworksACM Journal on Emerging Technologies in Computing Systems, 2017
- Unified particle physics for real-time applicationsACM Transactions on Graphics, 2014
- A method for registration of 3-D shapesIEEE Transactions on Pattern Analysis and Machine Intelligence, 1992
- Principal component analysis in linear systems: Controllability, observability, and model reductionIEEE Transactions on Automatic Control, 1981