CPG-based circuitry for controlling musculoskeletal model of human locomotor system

Andrii Dmytrovych Shachykov (1, 2, 3) , Patrick Henaff (2) , Anton Popov (1) , Olexandr Petrovych Shulyak (1)

1 National Technical University of Ukraine "Kyiv Polytechnic Institute" [Kiev]
2 NEUROSYS - Analysis and modeling of neural systems by a system neuroscience approach
3 UL - Université de Lorraine

In this paper, a new neuro-musculoskeletal simulator of human locomotor system is presented. This simulator is dedicated to reproduce healthy or altered walking gaits. It contains three joints per leg (hip, knee, ankle) controlled by twelve human muscle models activated by six specific models of central pattern generator (CPG). The CPG consists of three layers and four types of neurons and controls human leg joints. The CPGs are able to generate variable rhythmic signals by changing their intrinsic neural parameters which are controlled by descending signals from mesencephalic locomotor region (MLR), while output signals of motoneurons of CPGs control muscle models. Simulation results in Matlab show that it is possible to generate different stable walking gaits by changing intrinsic parameters of CPGs. According to these changes, the simulator can exhibit coherent or incoherent coordination between the two legs and consequently, stable or unstable walking gaits starting from the double support phase. Results show that this simulator will allow to reproduce walking gaits altered by basal ganglia decision-making system affected by Parkinson's disease.

Mots clés: central pattern generator; Parkinson's disease; musculoskeletal model; human walking

Domaines: Informatique [cs]; Rhumatologie et système ostéo-articulaire

Andrii Dmytrovych Shachykov, Patrick Henaff, Anton Popov, Olexandr Petrovych Shulyak. CPG-based circuitry for controlling musculoskeletal model of human locomotor system. BioCAS 2017 - IEEE Biomedical Circuits and Systems Conference, Oct 2017, Turin, Italy. ⟨hal-01798605⟩