Musculoskeletal system control method and system based on speed precision balance and device

Abstract

The invention belongs to the technical field of control, particularly relates to a musculoskeletal system control method and system based on speed precision balance and a device, and aims to solve the problem that an existing humanoid musculoskeletal robot control method cannot well perform antagonistic muscle collaborative contraction control. The method comprises the following steps of acquiring estimated motion precision of a musculoskeletal system through a Fitts' law; calculating a supervised item torque based on the estimated motion precision through a striatum inspired speed modulation strategy; calculating a muscle activation signal vector through a muscle activation signal network; and calculating an action award based on the muscle activation signal vector and the supervised item torque, then calculating a loss function, adjusting parameters of the muscle activation signal network based on the loss function, increasing the value of the action reward, and repeatedly iterating to obtain a muscle activation signal sequence required by control. According to the method, structural information of the musculoskeletal system is utilized, a universal antagonistic muscle cooperative contraction control strategy is constructed, and smooth movement is guaranteed.

Description

technical field [0001] The invention belongs to the technical field of control, and in particular relates to a musculoskeletal system control method, system and device based on trade-off of speed and accuracy. Background technique [0002] The adaptability of organisms allows them to flexibly adjust and execute behaviors, allowing learned and skilled movements to change according to the environment and task requirements. One of the typical strategies to achieve motion variability is the speed-accuracy trade-off, which reflects the trade-off between fast and accurate motion. How to implement such a flexible behavior strategy in humanoid musculoskeletal robots, so that the robot can generate general adaptability to the environment and tasks, is an attractive challenge. On the other hand, for humanoid musculoskeletal robot systems, the number of muscles is generally much larger than the number of joints. Redundant muscles not only bring difficulties to motor learning, but also...

AI Technical Summary

Problems solved by technology

[0003] In order to solve the above-mentioned problems in the prior art, that is, the existing humanoid musculoskeletal robot control method cannot well control the synergistic contraction of antagonistic muscles, the present invention provides a musculoskeletal system control based on speed precision trade-off method, said method comprising:

Images