Deep Learning Models for Stable Gait Prediction Applied to Exoskeleton Reference Trajectories for Children With Cerebral Palsy

Abstract

Gait trajectory prediction models have several applications in exoskeleton control; they can be used as feed-forward input to low-level controllers and to generate reference/target trajectories for position-controlled exoskeletons. In our study, we implement four deep learning models (LSTM, FCN, CNN and Transformer) that perform one-step-ahead gait trajectory prediction after training on gait patterns of typically developing children. We propose a methodology that optimises for stability in long-term forecasts, and evaluate the performance of the models on typically developing (TD) and Cerebral Palsy (CP) gait during recursive prediction of 200 time-steps in the future (which may lead to propagation of errors) and in the presence of varying levels of Gaussian noise (1%-5%). Results on TD gait show that the FCN and Transformer, with mean absolute errors (MAEs) for one-step-ahead predictions between 1.17°-1.63°, are the most suitable for the intended application. We also proposed an approach for generating adaptive trajectories that can be used as reference trajectories for position-controlled exoskeletons. Gait patterns from children with Cerebral Palsy were fed into gait trajectory prediction models trained on typically developing gait only, to generate corrective patterns. Preliminary results show that the gait patterns of typically developing children were introduced onto the generated trajectories.

Effect of Gait Speed on Trajectory Prediction Using Deep Learning Models for Exoskeleton Applications

“…In a previous study of ours, the fully connected neural network (FCNN) demonstrated low errors in both short-term and long-term gait trajectory prediction tasks and exhibited higher robustness to added noise [ 33 ]. These were the reasons for selecting the FCNN for this study.…”

“…The models were trained for 70 epochs, with training being terminated earlier (using the early stopping method) if the DTW distances on the validation set did not decrease for 20 epochs. In our previous study [ 33 ], we elaborated on how DTW distances are used to optimise gait trajectory prediction models.…”

A Transformer-Based Neural Network for Gait Prediction in Lower Limb Exoskeleton Robots Using Plantar Force

“…Utilizing the angle of hip and knee joints and plantar pressure data, Wu et al used a graph convolutional network model to classify the gait phase for a lower limb exoskeleton robot [ 27 ]. For children with cerebral palsy, Kolaghassi et al implemented four deep learning models (LSTM, FCN, CNN and a transformer) to predict joint angles and proposed an approach for adaptive trajectory generation based on a dataset consisting of flexion–extension angles of the hip, knee and ankle in the sagittal plane [ 28 ]. Huang et al attached an intelligent inertial measurement unit (IMU) to the surface of a shoe to sample the acceleration data of foot movements and proposed an online detection algorithm to identify the gait phase [ 29 ].…”

3D gait analysis in children using wearable sensors: feasibility of predicting joint kinematics and kinetics with personalized machine learning models and inertial measurement units

“…The primary focus of existing ML models for 3DGA in children lies in gait classification ( Kamruzzaman and Begg, 2006 ; Zhang et al, 2009 ; Zhang and Ma, 2019 ; Choisne et al, 2020 ; Khaksar et al, 2021 ) rather than the development of models for predicting gait time series. There are only a handful of studies focused on predicting children’s gait using ML techniques ( Kwon et al, 2012 ; Vigneron et al, 2017 ; Morbidoni et al, 2021 ; Kolaghassi et al, 2022 ; Kolaghassi et al, 2023 ). A research group used EMG sensors’ signals to predict children with cerebral palsy (CP) knee moment and achieved high correlation coefficients between 0.71 and 0.93 for different participants ( Kwon et al, 2012 ).…”

“…Another study proved the feasibility of using neural networks in predicting gait events from surface EMG signals in hemiplegic cerebral palsy ( Morbidoni et al, 2021 ). Other studies have employed ML techniques to estimate one-step-ahead gait trajectories to control lower-limb robotic devices in children with CP ( Kolaghassi et al, 2022 ; Kolaghassi et al, 2023 ). However, none of the mentioned studies utilized IMUs’ data to develop the ML model.…”