Dataset Information

A SE-DenseNet-LSTM model for locomotion mode recognition in lower limb exoskeleton.

ABSTRACT: Locomotion mode recognition in humans is fundamental for flexible control in wearable-powered exoskeleton robots. This article proposes a hybrid model that combines a dense convolutional network (DenseNet) and long short-term memory (LSTM) with a channel attention mechanism (SENet) for locomotion mode recognition. DenseNet can automatically extract deep-level features from data, while LSTM effectively captures long-dependent information in time series. To evaluate the validity of the hybrid model, inertial measurement units (IMUs) and pressure sensors were used to obtain motion data from 15 subjects. Five locomotion modes were tested for the hybrid model, such as level ground walking, stair ascending, stair descending, ramp ascending, and ramp descending. Furthermore, the data features of the ramp were inconspicuous, leading to large recognition errors. To address this challenge, the SENet module was incorporated, which improved recognition rates to some extent. The proposed model automatically extracted the features and achieved an average recognition rate of 97.93%. Compared with known algorithms, the proposed model has substantial recognition results and robustness. This work holds promising potential for applications such as limb support and weight bearing.

SUBMITTER: Tang J

PROVIDER: S-EPMC10909223 | biostudies-literature | 2024

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

A SE-DenseNet-LSTM model for locomotion mode recognition in lower limb exoskeleton.

Tang Jing J Zhao Lun L Wu Minghu M Jiang Zequan Z Cao Jiaxun J Bao Xiang X

PeerJ. Computer science 20240229

Locomotion mode recognition in humans is fundamental for flexible control in wearable-powered exoskeleton robots. This article proposes a hybrid model that combines a dense convolutional network (DenseNet) and long short-term memory (LSTM) with a channel attention mechanism (SENet) for locomotion mode recognition. DenseNet can automatically extract deep-level features from data, while LSTM effectively captures long-dependent information in time series. To evaluate the validity of the hybrid mode ...[more]

PMID: 38435551

Similar Datasets

Project description:Few studies have systematically investigated robust controllers for lower limb rehabilitation exoskeletons (LLREs) that can safely and effectively assist users with a variety of neuromuscular disorders to walk with full autonomy. One of the key challenges for developing such a robust controller is to handle different degrees of uncertain human-exoskeleton interaction forces from the patients. Consequently, conventional walking controllers either are patient-condition specific or involve tuning of many control parameters, which could behave unreliably and even fail to maintain balance. We present a novel, deep neural network, reinforcement learning-based robust controller for a LLRE based on a decoupled offline human-exoskeleton simulation training with three independent networks, which aims to provide reliable walking assistance against various and uncertain human-exoskeleton interaction forces. The exoskeleton controller is driven by a neural network control policy that acts on a stream of the LLRE's proprioceptive signals, including joint kinematic states, and subsequently predicts real-time position control targets for the actuated joints. To handle uncertain human interaction forces, the control policy is trained intentionally with an integrated human musculoskeletal model and realistic human-exoskeleton interaction forces. Two other neural networks are connected with the control policy network to predict the interaction forces and muscle coordination. To further increase the robustness of the control policy to different human conditions, we employ domain randomization during training that includes not only randomization of exoskeleton dynamics properties but, more importantly, randomization of human muscle strength to simulate the variability of the patient's disability. Through this decoupled deep reinforcement learning framework, the trained controller of LLREs is able to provide reliable walking assistance to patients with different degrees of neuromuscular disorders without any control parameter tuning. A universal, RL-based walking controller is trained and virtually tested on a LLRE system to verify its effectiveness and robustness in assisting users with different disabilities such as passive muscles (quadriplegic), muscle weakness, or hemiplegic conditions without any control parameter tuning. Analysis of the RMSE for joint tracking, CoP-based stability, and gait symmetry shows the effectiveness of the controller. An ablation study also demonstrates the strong robustness of the control policy under large exoskeleton dynamic property ranges and various human-exoskeleton interaction forces. The decoupled network structure allows us to isolate the LLRE control policy network for testing and sim-to-real transfer since it uses only proprioception information of the LLRE (joint sensory state) as the input. Furthermore, the controller is shown to be able to handle different patient conditions without the need for patient-specific control parameter tuning.

Project description:Walking rehabilitation processes include many repetitions of the same physical movements in order to replicate, as close as possible, the normal gait trajectories, and kinematics of all leg joints. In these conventional therapies, the therapist's ability to discover patient's limitations-and gradually reduce them-is key to the success of the therapy. Lower-limb robotic exoskeletons have strong deficiencies in this respect as compared to an experienced therapist. Most of the currently available robotic solutions are not able to properly adapt their trajectories to the biomechanical limitations of the patient. With this in mind, much research and development is still required in order to improve artificial human-like walking patterns sufficiently for valuable clinical use. The work herein reported develops and presents a method to acquire and saliently analyze subject-specific gait data while the subject dons a passive lower-limb exoskeleton. Furthermore, the method can generate adjustable, yet subject-specific, kinematic gait trajectories useful in programming controllers for future robotic rehabilitation protocols. A human-user study with ten healthy subjects provides the experimental setup to validate the proposed method. The experimental protocol consists in capturing kinematic data while subjects walk, with the donned H2 lower-limb exoskeleton, across three experimental conditions: walking with three different pre-determined step lengths marked on a lane. The captured ankle trajectories in the sagittal plane were found by normalizing all trials of each test from one heel strike to the next heel strike independent of the specific gait features of each individual. Prior literature suggests analyzing gait in phases. A preliminary data analysis, however, suggests that there exist six key events of the gait cycle, events that can adequately characterize gait for the purposes required of robotic rehabilitation including gait analysis and reference trajectory generation. Defining the ankle as an end effector and the hip as the origin of the coordinate frame and basing the linear regression calculations only on the six key events, i.e., Heel Strike, Toe Off, Pre-Swing, Initial Swing, Mid-Swing, and Terminal Swing, it is possible to generate a new calculated ankle trajectory with an arbitrary step length. The Leave-One-Out Cross Validation algorithm was used to estimate the fitting error of the calculated trajectory vs. the characteristic captured trajectory per subject, showing a fidelity average value of 95.2, 96.1, and 97.2%, respectively, for each step-length trial including all subjects. This research presents method to capture ankle trajectories from subjects and generate human-like ankle trajectories that could be scaled and computed on-line, could be adjusted to different gait scenarios, and could be used not only to generate reference trajectories for gait controllers, but also as an accurate and salient benchmark to test the human likeness of gait trajectories employed by existing robotic exoskeletal devices.

Dataset Information

A SE-DenseNet-LSTM model for locomotion mode recognition in lower limb exoskeleton.

Publications

A SE-DenseNet-LSTM model for locomotion mode recognition in lower limb exoskeleton.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets