Abstract:In factories, human posture recognition facilitates human–machine collaboration, human risk management, and workflow improvement. Compared to optical sensors, inertial sensors have the advantages of portability and resistance to obstruction, making them suitable for factories. However, existing product-level inertial sensing solutions are generally expensive. This paper proposes a low-cost human motion capture system based on BMI 160, a type of six-axis inertial measurement unit (IMU). Based on WIFI communication, the collected data are processed to obtain the displacement of human joints' rotation angles around XYZ directions and the displacement in XYZ directions, then the human skeleton hierarchical relationship was combined to calculate the real-time human posture. Furthermore, the digital human model was been established on Unity3D to synchronously visualize and present human movements. We simulated assembly operations in a virtual reality environment for human posture data collection and posture recognition experiments. Six inertial sensors were placed on the chest, waist, knee joints, and ankle joints of both legs. There were 16,067 labeled samples obtained for posture recognition model training, and the accumulated displacement and the rotation angle of six joints in the three directions were used as input features. The bi-directional long short-term memory (BiLSTM) model was used to identify seven common operation postures: standing, slightly bending, deep bending, half-squatting, squatting, sitting, and supine, with an average accuracy of 98.24%. According to the experiment result, the proposed method could be used to develop a low-cost and effective solution to human posture recognition for factory operation.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to develop a low - cost inertial measurement unit (IMU) motion capture system for the collection and recognition of operating postures. Specifically: 1. **High - cost problem**: Existing product - level inertial motion - capture devices are usually expensive, which limits their wide application in practical scenarios such as factories. This paper proposes a low - cost solution based on the BMI160 six - axis inertial measurement unit (IMU), aiming to reduce hardware costs. 2. **Operating posture recognition**: In the factory environment, human posture recognition is helpful for human - machine collaboration, risk management and workflow optimization. Compared with optical sensors, inertial sensors have the advantages of portability and anti - occlusion, so they are more suitable for the factory environment. In this paper, the BiLSTM model is used to recognize common operating postures, including seven postures: standing, slightly bending, deeply bending, half - squatting, full - squatting, sitting and supine. 3. **Real - time visualization**: In order to better display and analyze human motion, this paper establishes a digital human model on the Unity3D platform, and synchronously displays human motion through WiFi communication, realizing real - time visualization in the virtual reality environment. ### Main contributions - A low - cost motion - capture system based on BMI160 IMU is proposed, and the total hardware cost is only $3.60. - The BiLSTM model is used to accurately recognize seven common operating postures, with an average accuracy rate of 98.24%. - The feasibility and effectiveness of the system are verified through experiments, and its application potential in the virtual reality environment is demonstrated. ### Experimental design - **Hardware configuration**: Use the BMI160 IMU module, ESP8266 - NodeMCU communication module and IP5306 BMS charging module to build a tracker. - **Data processing**: Filter and drift - compensate the IMU signal through the Kalman filtering algorithm to ensure the accuracy and stability of the data. - **Posture recognition**: Based on the BiLSTM model, classify the collected joint displacement and angle data to realize the recognition of human postures. - **Real - time visualization**: Establish a digital human model on the Unity3D platform, and realize the real - time synchronous display of human motion through the Open Sound Control (OSC) protocol. Through the above methods, this paper provides an economical and efficient solution that can accurately collect and recognize human operating postures in the factory environment.

A Low-Cost Inertial Measurement Unit Motion Capture System for Operation Posture Collection and Recognition

Mpose: Environment- and Subject-Agnostic 3D Skeleton Posture Reconstruction Leveraging a Single Mmwave Device

Human-centric Application in Cyber-Physical System: an Inertial-based Motion Capture and Recognition System

Construction posture recognition with primitive joints extended planar normal vector quaternions

A Scalable and Wearable Self-Sensing IMU Sensor Network for Personalized Human Motion and Deformation Capture

PressInPose: Integrating Pressure and Inertial Sensors for Full-Body Pose Estimation in Activities

Real-Time Human Motion Capture Based on Wearable Inertial Sensor Networks

A Privacy-Preserving and Unobtrusive Sitting Posture Recognition System Via Pressure Array Sensor and Infrared Array Sensor for Office Workers

Human Posture Tracking System for Industrial Process Design and Assessment

An Attention-Based Deep Learning Approach for Inertial Motion Recognition and Estimation in Human-Robot Collaboration

Human Motion Capture and Recognition Based on Sparse Inertial Sensor.

Human Posture Transition-Time Detection Based upon Inertial Measurement Unit and Long Short-Term Memory Neural Networks

Training Data Selection and Optimal Sensor Placement for Deep-Learning-Based Sparse Inertial Sensor Human Posture Reconstruction

Design of Human Motion Acquisition System Based on MIMU

Classification of static postures with wearable sensors mounted on loose clothing

Design and implementation of a human posture capture system based on AHRS

View independent human posture identification using Kinect

Real-Time Onboard Human Motion Recognition Based On Inertial Measurement Units

Ultra Inertial Poser: Scalable Motion Capture and Tracking from Sparse Inertial Sensors and Ultra-Wideband Ranging

A Real-Time Posture Assessment System Based on Motion Capture Data for Manual Maintenance and Assembly Processes

A Novel Method for User-Defined Human Posture Recognition Using Kinect