Wearable Triboelectric Sensors Enabled Gait Analysis and Waist Motion Capture for IoT-Based Smart Healthcare Applications

Abstract

Gait and waist motions always contain massive personnel information and it is feasible to extract these data via wearable electronics for identification and healthcare based on the Internet of Things (IoT). There also remains a demand to develop a cost-effective human-machine interface to enhance the immersion during the long-term rehabilitation. Meanwhile, triboelectric nanogenerator (TENG) revealing its merits in both wearable electronics and IoT tends to be a possible solution. Herein, the authors present wearable TENG-based devices for gait analysis and waist motion capture to enhance the intelligence and performance of the lower-limb and waist rehabilitation. Four triboelectric sensors are equidistantly sewed onto a fabric belt to recognize the waist motion, enabling the real-time robotic manipulation and virtual game for immersion-enhanced waist training. The insole equipped with two TENG sensors is designed for walking status detection and a 98.4% identification accuracy for five different humans aiming at rehabilitation plan selection is achieved by leveraging machine learning technology to further analyze the signals. Through a lower-limb rehabilitation robot, the authors demonstrate that the sensory system performs well in user recognition, motion monitoring, as well as robot and gaming-aided training, showing its potential in IoT-based smart healthcare applications.

Keywords: human-machine interface; machine learning; robot-aided rehabilitation; smart healthcare; triboelectric sensors.

Figure 1

Schematics of the AIoT‐based smart healthcare. a) Wearable electronics including proposed TENG sensors for smart healthcare. b) Schematics of the AIoT‐based smart rehabilitation system with gait detection and waist motion capture. c) Characteristics and future applications of the smart rehabilitation system using TENG sensors.

Figure 2

Characterizations of the textile‐based TENG sensor. a) Structure and b) working mechanism of the TENG sensor. c) Open‐circuit voltages of the TENG sensor with different compressing forces. d) Output characterizations of the TENG sensor with 100 N compressing force and 500 mm min⁻¹ compressing speed. e) Power curves of the TENG with 50 and 100 N compressing forces. f) Output voltages during the cyclic compressing test. g) Images of TENG sensor and its dipped‐in‐water status. h) Output characterization of the TENG after dipping in the water.

Figure 3

HMI enhanced waist rehabilitation via intelligent safety belt. a) Schematics and working mechanism of the intelligent safety belt with four TENG sensors marked as W‐TENG‐1 to W‐TENG‐4. The safety belt is fixed to the fixture similar tothe rigid bracket of the rehabilitation robot. b) Hardware and flowchart for real‐time vehicle control. c) Images of experimental setup using the safety belt with TENGs to control the vehicle. d) Schematics and h) real‐time signals to control the vehicle by W‐TENGs, where motions and related signals are marked by the same number.

Figure 4

TENG‐based smart insole for gait analysis. a) Structure of the insole with two TENG sensors marked as I‐TENG‐1 and I‐TENG‐2. b) Schematics of four states of a typical contact cycle and c) the corresponding signals of normal walking. The walking states and associated signals are marked with the same number. d) Voltages of different walking modes, where the foot is on the ground at the original state. e) Real‐time walking speed detection based on the signal frequency. f) Voltages of jumping, going upstairs, and going downstairs, where the state (i) and state (ii) for energy storage and landing during jumping, respectively. g) TENG‐based smart insole to detect Parkinson's symptoms and h) to detect fall. FOG represents freezing of gait.

Figure 5

Patient identification based on the smart insole and machine learning. a) Schematic diagrams of the patient recognition by smart insole and machine learning. b) Overview of the recognition system. c) Real‐time signals for different patients. d) Confusion map of the machine learning training result. e) 3D plots of the I‐TENG sensor outputs corresponding to five patients, where the five participants in (d) are marked as patient A to patient E.

Figure 6

Demonstration of the robot‐aided rehabilitation with triboelectric sensory system. a) Schematics of the robot‐aided rehabilitation process. b) Configuration of the lower‐limb rehabilitation robot (iReGo) integrated with the TENG sensory system. This robot possesses two motor‐driven wheels and a safety belt to support and guide the walking training. c) Diagrams of training plan selection based on patient recognition. d) Gaming‐enhanced waist training using TENG‐based safety belt for improving the enjoyment. e) Demonstration of rehabilitation robot control with TENG‐based safety belt to help patient walk. f) Diagrams to illustrate the walking training process.