Stability-guaranteed reinforcement learning for Model Predictive Control gain tuning with transfer learning in Multi-UAV Systems

Khan, Abdul Manan

Stability-guaranteed reinforcement learning for Model Predictive Control gain tuning with transfer learning in Multi-UAV Systems

Lists

Khan, Abdul Manan (2026) Stability-guaranteed reinforcement learning for Model Predictive Control gain tuning with transfer learning in Multi-UAV Systems. In: The 2nd International Conference on Smart Mobility and Logistics Ecosystems (SMILE 2026), 8-11 February 2026, Dhahran, Kingdom of Saudi Arabia,.

[thumbnail of KhanAM_Stability guaranteed reinforcement learning_VoR.pdf]

Preview

PDF
KhanAM_Stability guaranteed reinforcement learning_VoR.pdf - Published Version
Available under License Creative Commons Attribution Non-commercial No Derivatives.
Download (486kB) | Preview

Official URL: https://www.sciencedirect.com/science/article/pii/...

Abstract

Tuning Model Predictive Control (MPC) weight matrices remains a tedious process that demands considerable expertise. A method is presented that enables a reinforcement learning agent to adjust gains online while enforcing Lyapunov-based bounds guaranteeing that every proposed gain lies inside a provably stable region. The resulting adaptive controller maintains stability regardless of policy network outputs. Across four Unmanned Aerial Vehicle (UAV) platforms spanning a 200-fold mass range (27 g to 5.5 kg), tracking improvements of 22–27% are observed on an aggressive 3D figure-8 trajectory spanning ±4.0 m on all axes (X, Y, and Z). Position root mean square error (RMSE) is reduced from 0.45–0.55 m to 0.33–0.43 m across all platforms, with variance reductions of 28–33%. No stability violations occurred throughout 60 evaluation trials. A projection operator clips out-of-bounds gains before they reach the MPC solver, acting as a hard safety layer. Sequential transfer learning reduces per-platform training by 75%. These findings demonstrate that formal stability constraints and learning-based adaptation can coexist
effectively.

Item Type:	Conference or Workshop Item (Paper)
ISSN:	2352-1465
Page Range:	pp. 948-955
Identifier:	10.1016/j.trpro.2026.04.077
Keywords:	MPC ; Reinforcement Learning ; Lyapunov Stability ; UAV Control ; Transfer Learning
Subjects:	Computing
Date Deposited:	29 Sep 2026
Dates:	Date Publication status 1 January 2026 Accepted 11 February 2026 Presented 2026 Published
School, department or research centre:	School of Computing and Engineering
Keywords:	MPC ; Reinforcement Learning ; Lyapunov Stability ; UAV Control ; Transfer Learning
URI:	https://repository.uwl.ac.uk/id/eprint/15445
Sustainable Development Goals:	Goal 9: Industry, Innovation, and Infrastructure