Deep Learning for Human Locomotion Analysis in Lower-Limb Exoskeletons: A Comparative Study
This work addresses the need for accurate and adaptive control in wearable robotics to enhance mobility for individuals with impairments or augment performance, though it is incremental as it compares existing methods on a specific dataset.
This paper tackled the problem of classifying ground conditions and estimating terrain parameters for lower-limb exoskeletons by comparing eight deep neural network backbones, achieving high accuracy in terrain classification (0.94 ± 0.04) and precise estimations for ramp slope (1.95 ± 0.58°) and stair height (15.65 ± 7.40 mm).
Wearable robotics for lower-limb assistance have become a pivotal area of research, aiming to enhance mobility for individuals with physical impairments or augment the performance of able-bodied users. Accurate and adaptive control systems are essential to ensure seamless interaction between the wearer and the robotic device, particularly when navigating diverse and dynamic terrains. Despite the recent advances in neural networks for time series analysis, no attempts have been directed towards the classification of ground conditions, categorized into five classes and subsequently determining the ramp's slope and stair's height. In this respect, this paper presents an experimental comparison between eight deep neural network backbones to predict high-level locomotion parameters across diverse terrains. All the models are trained on the publicly available CAMARGO 2021 dataset. IMU-only data equally or outperformed IMU+EMG inputs, promoting a cost-effective and efficient design. Indeeds, using three IMU sensors, the LSTM achieved high terrain classification accuracy (0.94 +- 0.04) and precise ramp slope (1.95 +- 0.58°) and the CNN-LSTM a stair height (15.65 +- 7.40 mm) estimations. As a further contribution, SHAP analysis justified sensor reduction without performance loss, ensuring a lightweight setup. The system operates with ~2 ms inference time, supporting real-time applications. The code is code available at https://github.com/cosbidev/Human-Locomotion-Identification.