Robot Finger Control via EMG Kernel Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies struggle to accurately detect and interpret user intentions to move robot fingers based on Electromyography (EMG) signals, particularly for applications involving upper limb amputees who lack external sensors.
Innovation Solution
A system utilizing a semi-unsupervised learning algorithm that applies a kernel matrix to multichannel EMG signals to determine mapping functions, allowing for the estimation of user intentions without the need for output sensors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If output sensors are used to detect finger movement, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent extracts and utilizes only the input EMG signals from the user's residual limb, completely removing the need for output sensors on the amputated hand. By focusing solely on the electrical signals from muscle activations in the remaining limb, the system achieves finger movement detection without requiring complex sensor arrays on the amputated portion.
Solution Approach 2:
The patent introduces an intermediary computational model that maps EMG signals from the residual limb to intended finger movements. This intermediary layer processes the electrical signals through machine learning algorithms to infer movement intentions, serving as a bridge between the available EMG data and the required movement control without needing direct sensor feedback from the amputated hand.
2Device complexity
If unsupervised learning algorithm is used, then device complexity is reduced, but measurement precision may worsen
Solution Approach 1:
The unsupervised learning algorithm performs self-service by automatically discovering patterns and mappings between EMG signals and finger movements without requiring pre-programmed training data or manual calibration. The system autonomously learns the relationship between muscle activation patterns and intended movements, eliminating the need for complex manual configuration while maintaining detection accuracy.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The system effectively detects user intentions to move robot fingers, enabling precise control of robot hands even in the absence of external sensors, thus improving interaction capabilities for upper limb amputees.
Implementation Method 1
a multichannel EMG signal acquisition device for EMG signal output using EMG electrodes
Data Source
AI summary
A system and a method for controlling a robot finger may include a processor to determine a mapping function using a kernel matrix means of a multichannel electromyographic (EMG) signal to train a learning algorithm using at least one EMG signal training sample), and a storage to store data and an algorithm driven by the processor. The processor determines a kernel matrix by applying a polynomial function of the second order to the multichannel EMG signal time sample, and performs an operation for the kernel matrix and the mapping function to output a signal for controlling the robot finger.