A sitting posture human motion monitoring system for a passenger-carrying environment

Abstract

This invention provides a seated human motion monitoring system for load-bearing environments, comprising: a PC, a multi-degree-of-freedom vibration platform, a data acquisition module, an algorithm module, a storage module, and a data display module. The algorithm module includes a data monitoring unit, a data separation unit, a data judgment unit, and a data decision unit. The data acquisition module can identify the spine of a seated human based on acquired skeletal geometric parameters, numerically representing its motion posture. The algorithm module can achieve sensor data fusion, locate skeletal points within non-ideal motion ranges, and realize differentiated calibration of skeletal regions. This invention can accurately quantify the motion posture of drivers and passengers under load-bearing vibration environments.

Description

Technical Field

[0001] This invention relates to the field of intelligent monitoring, and in particular to a seated human motion monitoring system designed for various environments. Background Technology

[0002] Vehicles such as cars, high-speed trains, and airplanes cause vibrations in the passenger environment during operation. The vibration frequency, amplitude, and direction have varying degrees of impact on the motion characteristics of drivers and passengers. To further quantify and assess human motion posture in the passenger environment, motion capture analysis has become a current research focus.

[0003] Passengers in a passenger environment can be categorized into three postures: standing, sitting, and lying down. The sitting posture, due to its broader application in real-world passenger environments, presents greater analytical demands. Existing motion capture analysis for sitting postures primarily uses sensors on the body surface for data acquisition, focusing on the back, buttocks, and abdomen to characterize the sitting posture. However, surface data is susceptible to the elasticity of muscle tissue, resulting in a less accurate representation of internal structural changes caused by vibration. Furthermore, the mapping characteristics of surface data may differ from the internal skeletal movements, random deformations, and coupling patterns under vibration. Additionally, existing motion capture technologies suffer from overlapping limb data, leading to inaccuracies, particularly in the spine where numerous coupled skeletal elements present significant data overlap issues. Therefore, there is an urgent need to develop a multi-source information fusion and multi-method collaborative monitoring system for analyzing the vibration characteristics of sitting humans in a passenger environment to comprehensively and effectively assess their posture. Summary of the Invention

[0004] This invention provides a seated human motion monitoring system for passenger environments to improve the level of seated human motion monitoring in passenger environments.

[0005] This invention provides a seated human motion monitoring system for load-bearing environments, comprising: a PC terminal, a multi-degree-of-freedom vibration platform, a data acquisition module, an algorithm module, a storage module, and a data display module;

[0006] The PC is connected to the data acquisition module, the storage module, the data display module, and the algorithm module, respectively; the multi-degree-of-freedom vibration platform is connected to the data acquisition module.

[0007] The data acquisition module extracts the skeletal geometric parameters of the test subject based on the human skeletal characteristics and transmits the skeletal geometric parameters to the PC terminal; the PC terminal is used to store, process, and display human motion data, and transmit the human skeletal geometric parameters to the internal algorithm module;

[0008] The algorithm module includes a data separation unit, a data judgment unit, a data monitoring unit, and a data decision unit; the data monitoring unit, the data separation unit, the data judgment unit, and the data decision unit are all connected to the algorithm module; the data separation unit is used to separate the X, Y, and Z axis data of each skeletal point of the human body transmitted to the PC; the data judgment unit is used to identify and judge the sensor data; the data monitoring unit is used to pre-calibrate skeletal points in non-ideal motion value ranges to achieve key monitoring; the data decision unit is used to analyze the data of the key monitored skeletal points and set decision-making concepts;

[0009] The multi-degree-of-freedom vibration platform is used to simulate complex vehicle vibration environments; the storage module is used to store data; the data display module is used to display the data differences of various human skeletal points, with a focus on displaying data from pre-calibrated and key monitoring areas.

[0010] The PC and the data acquisition module transmit control signals and sensing signals through Kinect 2.0 SDK, PoE switch, USB 3.0, and TCP / IP communication protocols.

[0011] Preferably, the data acquisition module includes a Kinect sensor and an optical motion capture system. The Kinect sensor collects and analyzes motion data of various parts of the human body through its own Exemplar system and machine learning algorithms to generate human skeletal structure lines, realizing motion tracking of skeletal nodes such as the head, shoulders, arms, torso, legs, and feet. The reflective markers of the optical motion capture system are fixed with reference to the positions of the 24 vertebrae of the real human spine, realizing motion capture of important parts of the spine such as the cervical, thoracic, and lumbar vertebrae. By fitting the data collected by the Kinect sensor and the optical motion capture system, complete human skeletal motion capture is achieved, wherein the equipment layout is as follows: the Kinect sensor is placed to the side of the driver / passenger; the capture lens of the optical motion capture system is placed directly behind the driver / passenger.

[0012] Preferably, the experiment is further divided into a preliminary experiment and a formal experiment. In the preliminary experiment, the driver and passengers sit upright in the experimental seat with their heads upright, eyes looking straight ahead, and hands placed on the upper surface of their thighs. The installation and settings of the data acquisition module are the same as in the formal experiment, and data is collected at the same frequency. The purpose of the preliminary experiment is to determine the threshold required for data comparison in the formal experiment. The formal experiment is carried out after the threshold is determined in the preliminary experiment. The data separation unit processes all the data transmitted to the PC, and the processing steps are as follows: first, distinguish the data of each monitoring point, and then separate the X, Y, and Z axis data of each point.

[0013] Preferably, the analysis cycle of the data judgment unit is set to multiple time periods; the data judgment unit is used to make a preliminary judgment on the separation results of the data in the first time period. If the sensor data exceeds the threshold, it enters the data monitoring unit for processing; if the sensor data does not exceed the threshold, it enters the data decision unit for processing; the data monitoring unit determines the corresponding skeletal point and acquisition time according to the threshold judgment result in the first time period; if the data of the key monitored skeletal point does not exceed the threshold in multiple time periods, the data decision unit removes this skeletal point data as an error.

[0014] Preferably, the data decision-making unit visualizes decision information through the data display module; the data monitoring unit displays pre-calibrated key monitoring skeletal points through the data display module.

[0015] Preferably, the storage module is used to store the data acquired by the data acquisition module and to provide storage space for the operation of the PC, which may include a read-only memory (RAM) for storing instructions and a random access memory (ROM) for storing application data.

[0016] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0017] This invention provides a seated human motion monitoring system for load-bearing environments, comprising: a PC terminal, a multi-degree-of-freedom vibration platform, a data acquisition module, an algorithm module, a storage module, and a data display module; the PC terminal is connected to the data acquisition module, the storage module, the data display module, and the algorithm module; the multi-degree-of-freedom vibration platform is connected to the data acquisition module;

[0018] The data acquisition module extracts the skeletal geometric parameters of the subject based on human skeletal features and transmits these parameters to the PC. The PC stores, processes, and displays human motion data and transmits the skeletal geometric parameters to its internal algorithm module. The algorithm module includes a data separation unit, a data judgment unit, a data monitoring unit, and a data decision unit. The data monitoring unit, data separation unit, data judgment unit, and data decision unit are all connected to the algorithm module. The data separation unit separates the X, Y, and Z axis data of each skeletal point transmitted to the PC. The data judgment unit identifies... The system comprises the following components: a data monitoring unit for pre-calibrating skeletal points in non-ideal motion ranges for focused monitoring; a data decision unit for analyzing data from key monitored skeletal points and setting decision-making criteria; a multi-degree-of-freedom vibration platform for simulating complex vehicle vibration environments; a storage module for storing data; and a data display module for displaying data differences from various human skeletal points, with a focus on pre-calibrated and key monitored data. The PC and data acquisition module transmit control and sensor signals via Kinect 2.0 SDK, a PoE switch, USB 3.0, and TCP / IP communication protocols.

[0019] This invention provides a seated human motion monitoring system for load-bearing environments. The optical motion capture system compensates for the insufficient setting of spinal monitoring points by the Kinect sensor. The system also performs step-by-step data fusion through algorithmic fusion to accurately capture human motion posture with skeletal monitoring as the target, and quantifies human motion characteristics under load-bearing vibration environments. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be explained below:

[0021] Figure 1 This is a main structural block diagram provided in the embodiments of the present invention;

[0022] Figure 2 This is a structural block diagram of the data acquisition module provided in an embodiment of the present invention;

[0023] Figure 3 These are the main human skeletal points monitored in the embodiments of the present invention;

[0024] Figure 4 This is a flowchart of the algorithm module provided in an embodiment of the present invention. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the protection scope of the embodiments of this application.

[0026] To address the existing problems, this application provides a seated human motion monitoring system for load-bearing environments. Based on skeletal characteristics, it accurately identifies the seated human motion posture in a load-bearing vibration environment. Furthermore, it uses algorithms to calibrate bones in non-ideal value ranges, quantifying the motion characteristics of the human skeletal structure under vibration.

[0027] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0028] Figure 1 The main structural block diagram provided for the embodiments of the present invention includes: a PC terminal, a multi-degree-of-freedom vibration platform, a data acquisition module, an algorithm module, a storage module, and a data display module.

[0029] The PC is used to store, process, and display human motion data, and the algorithm module is written using Visual Studio 2019 software. The data acquisition module extracts the skeletal geometric parameters of the test subject on the multi-degree-of-freedom vibration platform based on human skeletal characteristics. The algorithm module integrates the data monitoring unit, the data separation unit, the data judgment unit, and the data decision unit. The data separation unit is used to separate the X, Y, and Z axis data of each skeletal point transmitted to the PC. The data judgment unit is used to identify and judge the sensor data. The data monitoring unit is used to pre-calibrate skeletal points in non-ideal motion value ranges to achieve key monitoring. The data decision unit is used to analyze the data of the key monitored skeletal points and set decision-making concepts.

[0030] The multi-degree-of-freedom vibration platform is used to simulate complex riding environments and provide a vibration environment for whole-body human vibration experiments; the storage module is used to store human motion data transmitted from the data acquisition module to the PC; the data display module is used to display the data changes of each human skeletal point, with a focus on displaying the data differences of pre-calibrated and key monitoring areas, and displaying the skeletal points in the final calibration non-ideal value range after data processing; the PC and the data acquisition module transmit control signals and sensing signals through Kinect 2.0 SDK, POE switch and USB 3.0 protocol, TCP / IP and other communication protocols.

[0031] Figure 2 This is a structural block diagram of a data acquisition module provided in an embodiment of the present invention. The data acquisition module is used to collect skeletal motion data of the test subject on the multi-degree-of-freedom vibration platform and transmit it to a host computer PC. The data acquisition module includes a Kinect sensor and an optical motion capture system; wherein, the Kinect sensor is a KinectV2.0 model, which integrates an RGB color camera, an infrared projector and a depth (infrared) camera, with a acquisition frequency of 15 or 30 frames per second (fps), a monitoring range of 0.5-4.5 meters, a maximum of 6 people tracked simultaneously, and can simultaneously track 25 skeletal nodes of the human body; the optical motion capture system mainly consists of an optical motion capture lens, a calibration frame, reflective markers, a PoE switch, and a lens fixing device, which captures the displacement signals of reflective markers on the surface of the test subject by placing multiple optical motion capture lenses in the experimental area;

[0032] The Kinect V2.0 sensor is placed to the side of the multi-degree-of-freedom vibration platform and connected to the host PC via a data converter and USB 3.0 cable. The driver is then automatically installed. Afterward, the sensor needs to be self-tested, relevant parameters set, and the sensor turned on via the Kinect 2.0 SDK. The optical motion capture lens in the optical motion capture system is positioned behind the multi-degree-of-freedom vibration platform and facing the back of the subject using a tripod, clamps, and other fixing devices to ensure that the lens's field of view covers the capture area. During use, the lens is connected to the PoE switch via a network cable. The PoE switch powers the lens and transmits data and control signals via the TCP / IP protocol. The PoE switch is connected to the host PC via a network cable. After the system soft connection is successful, the L-shaped calibration frame is placed in the center of the experimental area, ensuring that the side with the reflective markers faces upward, to facilitate the calibration of the spatial coordinate system of the optical motion capture system.

[0033] Figure 3The main human skeletal points monitored in this embodiment of the invention are as follows. The KinectV2.0 sensor can simultaneously track 25 human skeletal nodes, including: head joint points, neck joint points, shoulder and neck joint points, mid-spine joint points, spinal base joint points, left shoulder joint points, left elbow joint points, left wrist joint points, left hand joint points, left thumb joint points, left palm joint points, left hip joint points, left knee joint points, left ankle joint points, left foot joint points, right shoulder joint points, right elbow joint points, right wrist joint points, right hand joint points, right thumb joint points, right palm joint points, right hip joint points, right knee joint points, right ankle joint points, and right foot joint points. The 24 reflective marker points of the optical motion capture system are corresponding to and pasted according to the positions of the actual spine C1-C7, T1-T12, and L1-L5. By fitting the data collected by the two devices, motion capture of the complete human skeleton is achieved.

[0034] In the preliminary experiment, the subjects sat upright on the experimental chair with their heads upright, eyes looking straight ahead, and hands placed on the upper surface of their thighs. The chair and lower legs should not touch. The arrangement, monitoring position, and acquisition frequency of the Kinect sensor and optical motion capture system were the same as in the formal experiment to determine the threshold required for data comparison in the formal experiment.

[0035] Figure 4The flowchart of the algorithm module provided in the embodiment of the present invention. The data separation unit processes the complete human motion data transmitted from the PC. First, it separates the data for each monitored skeletal point. Second, it separates the X, Y, and Z axial displacement data for each skeletal point. Finally, it performs frame-by-frame and multi-frame analysis on the separated data. The frame-by-frame analysis compares the displacement data between adjacent frames with a pre-determined displacement value change threshold. The multi-frame analysis statistically analyzes the displacement data within multiple frames and compares it with a pre-determined displacement value change threshold. The data judgment unit's analysis cycle is divided into multiple time periods. The length and number of time periods can be adjusted according to experimental requirements. The data judgment unit performs preliminary judgment on the frame-by-frame and multi-frame analysis data in the first time period. If the sensor data exceeds the threshold, it enters the data monitoring unit for processing; if the data does not exceed the threshold, it enters the data decision unit for processing. The data decision unit plots the displacement trend curves of each skeletal point based on the data transmitted after the data judgment unit's judgment of the first time period, analyzes the parts with more prominent curve changes and the acquisition time, and uses this information to perform secondary analysis on some skeletal points according to experimental requirements. The analysis involves point calibration or pre-calibration decisions. The data monitoring unit, based on the data record transmitted after the data judgment unit's judgment of the first time history, records the skeletal points and times to which data exceeding the threshold belong, performing pre-calibration processing and focusing on monitoring data changes at these skeletal points. Subsequently, data is transmitted to the data judgment unit multiple times over N time histories. If the data at the key monitored skeletal point does not exceed the threshold again, this skeletal point data is removed as an error and removed from the list of key monitored skeletal points. If the pre-calibrated key monitored skeletal point data exceeds the threshold again... The system will then make another judgment and input the data into the data decision unit for decision-making in two ways: If the data of the key monitored skeletal point exceeds the threshold a few times in the subsequent N time periods, the point will be removed or treated as a secondary analysis point according to the pre-experiment settings. Then, it will be analyzed to determine whether the skeletal point should be marked as a skeletal point in a non-ideal value range. If the data of the key monitored skeletal point exceeds the threshold multiple times in the subsequent N time periods, the point will be treated as an important analysis point, and the system will determine whether the skeletal point should be marked as a skeletal point in a non-ideal value range according to the experimental requirements.

[0036] The pre-calibrated skeletal points in the data monitoring unit, the displacement change curves of each skeletal point drawn by the data decision unit, the secondary analysis skeletal points, the important analysis skeletal points, and the skeletal points in the finally determined non-ideal numerical range will all be visualized through the data display module.

[0037] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A sitting human motion monitoring system for a bearer environment, characterized by The application relates to a human body motion data acquisition and processing system. The PC terminal is connected with the data acquisition module, the storage module, the data display module and the algorithm module; the multi-degree-of-freedom vibration platform is connected with the data acquisition module. The data acquisition module extracts the skeletal geometric parameters of the test personnel according to the human body skeletal characteristics and transmits the skeletal geometric parameters to the PC terminal; the PC terminal is used for storing, processing and displaying human body motion data and transmitting the human body skeletal geometric parameters to the internal algorithm module; The data acquisition module comprises a Kinect sensor and an optical motion capture system; the Kinect sensor is internally integrated with an RGB color camera, an infrared projector and a depth camera, the acquisition frequency is 15 or 30 frames per second, the monitoring range is 0.5-4.5 meters, the maximum number of people tracked simultaneously is 6, and 25 skeletal nodes of the human body can be tracked simultaneously; the Kinect sensor collects and analyzes human body motion data through the Exemplar system and the machine learning algorithm of the Kinect sensor, generates a human body skeleton line, and realizes motion tracking of the skeletal nodes of the head, shoulder, arm, trunk, leg and foot of the human body; the optical motion capture system mainly comprises optical motion capture lenses, a calibration frame, reflective marker points, a POE switch and a lens fixing device, and the reflective marker points of the optical motion capture system are fixed according to the positions of the 24 vertebrae of the real human spine, so as to realize motion capture of the cervical vertebrae, thoracic vertebrae and lumbar vertebrae; the acquisition data of the Kinect sensor and the optical motion capture system are fitted, and complete human body skeletal motion capture is realized, wherein the device layout is as follows: the Kinect sensor is arranged on the side of the driver; the capture lenses in the optical motion capture system are arranged behind the driver; The algorithm module comprises a data separation unit, a data judgment unit, a data monitoring unit and a data decision unit; the data monitoring unit, the data separation unit, the data judgment unit and the data decision unit are connected with the algorithm module; the data separation unit is used for separating the X, Y and Z axial data of each skeletal point position of the human body transmitted to the PC terminal; the data judgment unit is used for identifying and judging the sensing data; the data monitoring unit is used for pre-calibrating the skeletal point positions in the non-ideal motion value interval, so as to realize key monitoring; and the data decision unit is used for analyzing the skeletal point position data of the key monitoring and setting a decision idea; The analysis cycle of the data judgment unit is set as multiple time histories; the data judgment unit is used for preliminarily judging the data processed by the data separation unit in the first time history; if the sensing data exceeds a threshold value, the sensing data is processed in the data monitoring unit; if the sensing data does not exceed the threshold value, the sensing data is processed in the data decision unit. ​ The data decision unit draws displacement trend curves of each skeletal point according to the data transmitted by the data judgment unit after judging the first time history, analyzes the parts with relatively prominent curve changes and the collection time, and makes secondary analysis point calibration or pre-calibration decision for part of the skeletal points according to the experimental requirements; the data monitoring unit determines the corresponding skeletal point position and collection time according to the threshold judgment result in the first time history, and performs pre-calibration processing and focuses on monitoring the data changes of the skeletal point; thereafter, the data is transmitted into the data judgment unit multiple times in N time histories; if the data of the key monitoring skeletal point position does not exceed the threshold in multiple time histories, the data decision unit removes the skeletal point position data as an error; if the data of the pre-calibrated key monitoring skeletal point position exceeds the threshold again, the data is transmitted into the data decision unit for decision in two cases; if the data of the key monitoring skeletal point exceeds the threshold in the subsequent N time histories for a small number of times, the point position is removed or used as a secondary analysis point according to the setting before the experiment, and then the skeletal point position is analyzed and determined whether to be calibrated as a skeletal point position in a non-ideal numerical range; if the data of the key monitoring skeletal point exceeds the threshold multiple times in the subsequent N time histories, the point is used as an important analysis point, and whether to be calibrated as a skeletal point position in a non-ideal numerical range is determined according to the experimental requirements; The multi-degree-of-freedom vibration platform is used to simulate a complex passenger-carrying vibration environment; the storage module is used to store data; the data display module is used to display the data differences of each human skeletal point position, with the data of the pre-calibrated and key monitoring positions being highlighted. The PC end and the data acquisition module transmit control signals and sensing signals through Kinect2.0 SDK, POE switch, USB3.0, and TCP / IP communication protocol.

2. The sitting posture human motion monitoring system for a bearer- oriented environment according to claim 1, characterized by, The experiment is further divided into a pre-experiment and a formal experiment; in the pre-experiment, the driver and passenger sit on the experimental seat in an upright sitting position, with the head upright, the eyes straight ahead, and the hands placed on the upper surface of the thighs; the installation and setting of the data acquisition module are the same as those in the formal experiment, and data is collected at the same frequency; the purpose of the pre-experiment is to determine the threshold required for data comparison in the formal experiment, and the formal experiment is carried out after the threshold is determined in the pre-experiment; the data separation unit processes all data transmitted to the PC end, and the processing steps are as follows: first, the data of each monitoring point is distinguished, and then the X, Y, and Z axis data of each point are separated.

3. The sitting posture human motion monitoring system for a bearer- oriented environment according to claim 1, characterized by, The data decision unit realizes decision information visualization through the data display module; the data monitoring unit displays the pre-calibrated key monitoring skeletal point position through the data display module.

4. The sitting human motion monitoring system for a bearer- oriented environment according to any one of claims 1 to 3, characterized in that, The storage module is used to store the data acquired by the data acquisition module and provide storage space for the PC end operation, including a read-only memory (RAM) for storing instructions and a random read-write memory (ROM) for storing application data.

Images