Prediction of Lower Limb Exoskeleton Gait Trajectories Based on Plantar Pressure Information | Semantic Scholar (2025)

A transformer-based neural network comprising convolution and variational mode decomposition (VMD) to predict bilateral hip and knee joint angles utilizing the plantar pressure and reduces the mean absolute error and mean squared error compared to the CNN model, the transformer model and the CNN transformer model.

The method can obtain the deep spatial-temporal relationship of gait trajectories by training large gait data, which avoids the complex modeling of human body and lower-limb exoskeleton.

Experimental results indicate that this long-short term memory neural network with feature-level fusion can achieve a superior prediction performance compared with other traditional neural networks and the trained LSTM model also presents better adaptability between individuals.

A new algorithm model of BP neural network based on support vector machines (SVMBP) is proposed, which is more suitable for gait detection and recognition in wearing exoskeleton robots, and it avoids the shortcomings of SVM in practical applications.

This review intends to first explain the working principles of mainstream insole plantar sensing techniques and design considerations such as sensing material selection and electronics design requirements, and then the state‐of‐the‐art algorithms for plantar pressure distribution reconstruction.

New guidelines for the selection of plantar pressure sensor placements are provided and intelligent optimization algorithms are incorporated into new ways to improve the accuracy of wearable device analysis.

Various control methods and/or techniques used in 73 lower limb exoskeletons are analyzed, classifying their information in different sections such as trade names, biosignals used as information inputs through the mechatronic system, sensors that obtain information on control parameters such as position and force, actuators that perform the selected functions, and their common applications.

This review focuses on examining the most recent developments and problems associated with exoskeleton control systems, particularly during the last few years (2017–2022), and the trends and challenges of cooperative control, particularly multi-information fusion, are discussed.

This paper proposes to model the so-called aleatoric uncertainty of a differentiable (soft) version of DTW, which is the smallest weighted path distance among all paths, and the regularization term is the aggregate of log-variances along the path.

...

...