Laser-Induced Graphene Sensors for Wearable Gait Recognition

In a recent paper published in Microsystems & Nanoengineering, researchers integrated a pressure sensor array based on laser-induced graphene (LIG) and multiple sensor units into wearable gait recognition sensors or exoskeleton robots.

Study: Laser-Induced Graphene Sensors for Wearable Gait Recognition . Image Credit: Gorodenkoff/Shutterstock.com

Exoskeletons Robotics and Gait Recognition Technology

Exoskeleton robots have shown diverse applications in fields like freight transport, healthcare, disaster relief, and rehabilitation due to their ability to empower human capabilities. Specifically, lower-limb-assisted exoskeleton robots are the center of attraction as they can mimic the leg movements of the wearer and provide assistance in load-bearing.

Unlike industrial robotic arms, the lower limb exoskeleton robot’s motion trajectory is planned according to the wearer’s movement intentions and assistance requirements rather than being preprogrammed. Gait recognition is a prerequisite for human–robot cooperation involving lower limb exoskeleton robots.

To evaluate gait perception, wearable sensors are required to collect pressure, velocity, and acceleration data from the human foot. The development of flexible sensors, particularly through laser direct writing technology, has enabled the design of wearable sensors. Thus, LIG-based gait recognition sensor systems improve gait information accuracy or human-robot interaction and can potentially be used in rehabilitation medicines.

Fabrication of LIG-based Recognition Sensor

The researchers used LIG to fabricate the sensor via one-step laser ablation of polyamide films. LIG was transferred onto soft polymers like polydimethyl siloxane, ecoflex, and hydrogels creating composites for flexible devices.

For the laser ablation step, the Universal laser System VLS 350 under laser fluence from 4.3 to 6.91 J/cm² with a wavelength of 10.6 µm was used. To integrate the intelligent insole, the researchers screen-printed silver electrodes on polyamide and cut into the shape of an insole by laser. A specialized input signal amplification (ISA) unit was also designed for the insole, which converts and amplifies the resistance of the pressure sensor unit of all channels into a voltage signal.

Characterization and Data Analysis

Related Stories

The laser-textured LIG surface’s morphology was analyzed using a thermal field emission scanning electron microscope (FE-SEM) and a 3D measuring laser microscope. In addition, Raman spectra were obtained using a 532 nm wavelength. The pressure sensors’ electric current measurements were conducted with a digital multi-meter (Key sight, 34470A), while sensor pressure was dynamically measured using a compression testing machine.

Five healthy adults aged 23-31 participated in the study, with a sampling and control frequency of 100 Hz. The input signal amplifier’s (ISA) size was 20mm×24mm, with each module containing two amplification processing units. The authors used four pressure sensor units and amplified their signals.

The analog-to-digital converter (ADC) module, comprising a power supply and voltage acquisition module, collected and transmitted voltage signals to the sensor network via the controller area network (CAN) bus. It measured 24mm×50mm, had a 16-bit resolution, and simultaneously collected signals from eight channels within a range of -10V to 10V.

Gait recognition and exoskeleton robot test data were gathered using ISA and ADC, and data analysis was performed using MATLAB 2020b and a microcontroller unit (MCU). The study protocol was approved by the ethical committee of Zhejiang University’s College of Biomedical Engineering & Instrument Science.

Observations

The system comprises seven pressure sensor units strategically positioned at foot stress points to monitor plantar pressure changes and distribution during various gait phases. Each gait cycle involves stages such as initial contact (IC), loading response (LR), mid-stance (MS), terminal stance (TS), and swing (SW). The pressure sensors, constructed with three layers including PI film with LIG patterns, laser-textured LIG on a polydimethylsiloxane (PDMS) layer, and a poly(ethylene terephthalate) (PET) encapsulation layer, enable real-time identification of these gait phases.

The sensing mechanism involves changes in electric current due to variations in the contact area between the sensor components under pressure. The pressure sensor’s sensitivity was optimized through laser texturing, demonstrating increased performance. Additionally, the cycling stability and sensitivity of the flexible pressure sensor were extensively characterized, ensuring reliable and stable performance over thousands of cycles.

Multiple LIG pressure sensor units were integrated into an intelligent insole, which was embedded in the shoes of an exoskeleton robot to provide real-time feedback on plantar pressure. The hardware system, including ISA, ADC, and MCU, facilitated signal amplification, acquisition, and processing.

A gait recognition model, based on the support vector machine (SVM) algorithm, achieved high prediction accuracy, demonstrating the reliability of the LIG-based gait recognition sensor system. Additionally, real-time walking experiments verified the system’s effectiveness, even at different gait frequencies, highlighting its potential for practical application in exoskeleton robotics.

Conclusion

This work presents a wearable sensor system that supports an exoskeleton for the control of human–robot interaction and has a timely feedback gait phase function. A quick and tailored laser processing method led to the successful development of an embedded pressure sensor unit. The method makes it possible to fabricate the conductive LIG patterns on a wide scale consistently.

Through pressure mapping, the exoskeleton robot is integrated with many sensor units and a printed circuit board inserted in the insoles to reflect the wearer’s gait. The SVM-based recognition algorithm aims to achieve extremely precise gait recognition, with experimental findings demonstrating a 99.85% accuracy rate for the gait recognition sensor system and real-world exoskeleton applications confirming its dependability.

Journal Reference

Sun, M., Cui, S., Wang, Z., Luo, H., Yang, H., Ouyang, X., & Xu, K. (2024). A laser-engraved wearable gait recognition sensor system for exoskeleton robots. Microsystems & Nanoengineering, 10(1), 1-9. https://doi.org/10.1038/s41378-024-00680-x, https://www.nature.com/articles/s41378-024-00680-x