Abstract:Locomotion transition recognition with external disturbances is a key issue in lower limb robotic prostheses. Redonning the prosthetic socket and individual differences are encountered frequently in daily life. They change the feature distribution and further fail the previously trained recognition model. To bridge the technical gap between laboratory validation and practical demands, researchers need to construct an adaptive recognition method with high accuracy, quick response, time-efficient calibration, and minimal interference to the human body. Our study developed an adaptive recognition algorithm based on two miniaturized inertial sensors (anterior thigh and forefoot) and a foot pressure sensor, which can easily be integrated into the prosthesis. The algorithm fused probability-based fuzzy classifiers, which took heuristic-based features as inputs with a dynamic-time-warping-based automatic template generation block. It enables the recognition model to quickly fit the parameters for an untrained mode after external disturbances. We validated the proposed adaptive recognition method on 13 healthy subjects, performing 18 motion transitions with random interday intervals and across user evaluation, both offline and online. Even with the data of one subject as the prior knowledge, the other subjects produced an average accuracy of 98.04% and an average transition time latency of -180.99 ms without manual training in interday and intersubject uses. We also tested the method on two subjects using robotic prostheses, one person with a transfemoral amputation and the other person with a transtibial amputation. The average accuracy was still as high as 98.06% (transfemoral) and 100% (transtibial) without manual calibration after seven days of intervals. The results proved that the adaptive recognition method is robust to sensor redonning and user changes. Compared with the state-of-the-art methods, our study makes a further step toward the practical use of robotic prostheses. Future studies will be conducted to implement the algorithm on the robotic prosthesis for more extensive tests.

A Real-Time Intent Recognition System Based on SoC-FPGA for Robotic Transtibial Prosthesis.

Implementing a SoC-FPGA Based Acceleration System for On-Board SVM Training for Robotic Transtibial Prostheses.

Real-time Visual Static Hand Gesture Recognition System and Its FPGA-based Hardware Implementation

Design of Gait Recognition System Based on FPGA and DSP

A Real-Time FPGA-Based Gesture Recognition System

On-board Training Strategy for IMU-Based Real-Time Locomotion Recognition of Transtibial Amputees with Robotic Prostheses

Performance Analysis of Hardware Acceleration for Locomotion Mode Recognition in Robotic Prosthetic Control.

Design and Implementation of Embedded Real-Time English Speech Recognition System Based on Big Data Analysis

FPGA-based Gait Control System for Passive Bipedal Robot.

Co-design and Implementation of Image Recognition Based on ARM and FPGA

An SoC-based Platform for Integrated Multi-Axis Motion Control and Motor Drive

Design of prosthetic hand control system based on FPGA

Adaptive Locomotion Transition Recognition with Wearable Sensors for Lower Limb Robotic Prosthesis

Industrial robot arm controller based on programmable System-on-Chip device

Implementation of Joint Servo Controller on-Chip

Hardware Implementation of a Virtual Blind Cane on FPGA

Design and Implementation of Rehabilitation Training and Positioning System Based on Multi-Sensor Information Fusion

Real-time target tracking and positioning on FPGA

A Highly Integrated Joint Servo System Based on FPGA with Nios II Processor

A High Performance FPGA-based Joint Controller with Hardware/software Co-Design Method

FPGA Implementation of a CORDIC-Based Joint Angle Processor for a Climbing Robot