Residual Neural Terminal Constraint for MPC-based Collision Avoidance in Dynamic Environments

Bojan Derajic, Mohamed-Khalil Bouzidi, Sebastian Bernhard, Wolfgang Hönig
Proceedings of The 9th Conference on Robot Learning, PMLR 305:1452-1469, 2025.

Abstract

In this paper, we propose a hybrid MPC local planner that uses a learning-based approximation of a time-varying safe set, derived from local observations and applied as the MPC terminal constraint. This set can be represented as a zero-superlevel set of the value function computed via Hamilton-Jacobi (HJ) reachability analysis, which is infeasible in real-time. We exploit the property that the HJ value function can be expressed as a difference of the corresponding signed distance function (SDF) and a non-negative residual function. The residual component is modeled as a neural network with non-negative output and subtracted from the computed SDF, resulting in a real-time value function estimate that is at least as safe as the SDF by design. Additionally, we parametrize the neural residual by a hypernetwork to improve real-time performance and generalization properties. The proposed method is compared with three state-of-the-art methods in simulations and hardware experiments, achieving up to 30% higher success rates compared to the best baseline while requiring a similar computational effort and producing high-quality (low travel-time) solutions.

Cite this Paper

BibTeX
@InProceedings{pmlr-v305-derajic25a, title = {Residual Neural Terminal Constraint for MPC-based Collision Avoidance in Dynamic Environments}, author = {Derajic, Bojan and Bouzidi, Mohamed-Khalil and Bernhard, Sebastian and H\"{o}nig, Wolfgang}, booktitle = {Proceedings of The 9th Conference on Robot Learning}, pages = {1452--1469}, year = {2025}, editor = {Lim, Joseph and Song, Shuran and Park, Hae-Won}, volume = {305}, series = {Proceedings of Machine Learning Research}, month = {27--30 Sep}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v305/main/assets/derajic25a/derajic25a.pdf}, url = {https://proceedings.mlr.press/v305/derajic25a.html}, abstract = {In this paper, we propose a hybrid MPC local planner that uses a learning-based approximation of a time-varying safe set, derived from local observations and applied as the MPC terminal constraint. This set can be represented as a zero-superlevel set of the value function computed via Hamilton-Jacobi (HJ) reachability analysis, which is infeasible in real-time. We exploit the property that the HJ value function can be expressed as a difference of the corresponding signed distance function (SDF) and a non-negative residual function. The residual component is modeled as a neural network with non-negative output and subtracted from the computed SDF, resulting in a real-time value function estimate that is at least as safe as the SDF by design. Additionally, we parametrize the neural residual by a hypernetwork to improve real-time performance and generalization properties. The proposed method is compared with three state-of-the-art methods in simulations and hardware experiments, achieving up to 30% higher success rates compared to the best baseline while requiring a similar computational effort and producing high-quality (low travel-time) solutions.} }
Endnote
%0 Conference Paper %T Residual Neural Terminal Constraint for MPC-based Collision Avoidance in Dynamic Environments %A Bojan Derajic %A Mohamed-Khalil Bouzidi %A Sebastian Bernhard %A Wolfgang Hönig %B Proceedings of The 9th Conference on Robot Learning %C Proceedings of Machine Learning Research %D 2025 %E Joseph Lim %E Shuran Song %E Hae-Won Park %F pmlr-v305-derajic25a %I PMLR %P 1452--1469 %U https://proceedings.mlr.press/v305/derajic25a.html %V 305 %X In this paper, we propose a hybrid MPC local planner that uses a learning-based approximation of a time-varying safe set, derived from local observations and applied as the MPC terminal constraint. This set can be represented as a zero-superlevel set of the value function computed via Hamilton-Jacobi (HJ) reachability analysis, which is infeasible in real-time. We exploit the property that the HJ value function can be expressed as a difference of the corresponding signed distance function (SDF) and a non-negative residual function. The residual component is modeled as a neural network with non-negative output and subtracted from the computed SDF, resulting in a real-time value function estimate that is at least as safe as the SDF by design. Additionally, we parametrize the neural residual by a hypernetwork to improve real-time performance and generalization properties. The proposed method is compared with three state-of-the-art methods in simulations and hardware experiments, achieving up to 30% higher success rates compared to the best baseline while requiring a similar computational effort and producing high-quality (low travel-time) solutions.
APA
Derajic, B., Bouzidi, M., Bernhard, S. & Hönig, W.. (2025). Residual Neural Terminal Constraint for MPC-based Collision Avoidance in Dynamic Environments. Proceedings of The 9th Conference on Robot Learning, in Proceedings of Machine Learning Research 305:1452-1469 Available from https://proceedings.mlr.press/v305/derajic25a.html.

Related Material