Autonomous adaptive motor control with neuron firing pattern and muscle growth
Various models for developing motion control methods using neuromuscular models have been proposed in many fields, such as biophysics, medicine, robotics, computer graphics, computer vision, and so on. This article proposes an autonomous adaptive model, Firing Pattern Controller and Muscle Growth (FPCMG), whose foundation is an implementation of the growth of muscle fiber used in tandem with the firing patterns of the neurons controlling it. The FPCMG model then consists of a motoneuron controller model with interneuron firing patterns and a muscle fiber model in which both slow- and fast-twitch fibers develop with a reinforcement learning rule with delayed reward. It is applied to a simple musculoskeletal model to demonstrate its validity. It simulates an isometric contraction using one joint and two muscles. The system learns different firing patterns depending on muscular strength, and adapts to achieve the target contraction.
Authors
- Shun Ogawa (ORCID: https://orcid.org/0000-0001-8278-4299)
- Yuko Ishiwaka
Institutions
- SoftBank Group (Japan) (JP)
- SoftBank Corp. (Japan) (JP)
Publication Details
- Journal
- Scientific Reports
- Published
- 2026-11-14
- DOI
- https://doi.org/10.1038/s41598-026-79660-2
- Primary Topic
- Muscle activation and electromyography studies
- Type
- article
- Field-Weighted Citation Impact
- 0.00