Lower limb exoskeleton robots are increasingly used in various applications, requiring accurate gait recognition for effective control and human-robot interaction. Existing sensor systems often lack the flexibility and comfort needed for seamless integration. This research addresses this challenge by developing a wearable gait recognition system using flexible, reliable LIG-based pressure sensors. The use of laser direct writing technology allows for customized, high-precision fabrication of these sensors, overcoming limitations of traditional flexible sensor fabrication methods. The LIG-based sensors, integrated into a smart insole, are designed to accurately capture plantar pressure changes during different gait phases, providing crucial feedback for exoskeleton robot control and gait analysis applications, particularly in rehabilitation medicine.

The introduction reviews existing research on exoskeleton robots and their applications in freight transport, disaster relief, and rehabilitation. It highlights the importance of gait information for effective exoskeleton control and discusses limitations of traditional wearable sensor systems, emphasizing the need for flexible and comfortable solutions. The development of flexible electronics and various flexible sensors, such as pressure, temperature, and humidity sensors, for tracking human activities is reviewed, noting limitations due to batch-to-batch variations in fabrication. The suitability of laser direct writing and laser-induced graphene (LIG) for creating high-precision flexible sensors is discussed, referencing previous studies demonstrating the fabrication of various sensors using this technique.

The researchers designed and fabricated a wearable gait recognition sensor system based on LIG. The system consists of seven pressure sensor units strategically placed within an insole to capture plantar pressure at key points during a gait cycle. Each sensor unit comprises three layers: a PI film with LIG patterns, a laser-textured LIG/PDMS layer, and a PET encapsulation layer. The LIG patterns act as sensing elements, with pressure changes altering the contact area between the laser-textured LIG/PDMS and interdigital LIG electrodes, resulting in an electrical current change. The fabrication process involves laser ablation of PI films to create the LIG structures, followed by transferring the LIG onto PDMS and laser texturing to enhance sensitivity. Scanning electron microscopy (SEM) and Raman spectroscopy were used to characterize the LIG structure. The sensitivity of the sensors was evaluated by measuring the change in electric current under various pressures. The sensitivity was defined as S = (ΔI/I₀)/P, where I₀ is the initial current, P is the applied pressure, and ΔI is the change in current. Multiple sensor units were integrated into a flexible, wearable insole. The hardware system included a signal amplifier, an analog-to-digital converter (ADC), and a microcontroller unit (MCU), with data transmission via the controller area network (CAN). A machine learning algorithm was employed to improve gait recognition accuracy. The entire system was integrated with an exoskeleton robot for testing and validation.

The LIG-based pressure sensors demonstrated high sensitivity, with laser texturing significantly improving performance compared to untextured sensors. The sensors exhibited excellent long-term stability, showing minimal degradation after 5000 cycles. The integrated intelligent insole system achieved a gait recognition accuracy of 99.85%. The system successfully provided real-time gait information for exoskeleton robot control, demonstrating its effectiveness in human-robot interaction.

The high accuracy and stability of the LIG-based gait recognition system address the need for reliable gait information in exoskeleton robot control. The use of laser direct writing for fabrication offers advantages in terms of precision, efficiency, and customization. The system’s success demonstrates the potential of LIG-based sensors for use in wearable applications requiring high sensitivity and flexibility. The system’s potential for applications in rehabilitation medicine and gait analysis is highlighted.

This study successfully demonstrated a highly accurate and reliable wearable gait recognition system for exoskeleton robots using laser-engraved LIG-based pressure sensors. The system's high accuracy, stability, and ease of integration highlight its potential for improving human-robot interaction in various applications. Future research could focus on further miniaturization of the sensors, integration with other sensing modalities, and development of more sophisticated machine learning algorithms for improved gait recognition under diverse conditions.

While the system demonstrated high accuracy in controlled environments, further testing is needed to evaluate its performance under more realistic and varied conditions, including different walking speeds, terrains, and individual variations in gait patterns. The study could benefit from a larger sample size for participant testing to strengthen the generalizability of the results.