Hybrid Motion Estimation for VR Training

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional virtual training systems fail to accurately estimate the position and posture of trainees due to inaccuracies in depth map-based methods and divergent errors from motion sensors, making it difficult for precise training results and prolonged analysis in military training scenarios.

Innovation Solution

A system and method that combine motion sensors and depth sensors to estimate the position, posture, and traveled distance of trainees using a converging unit for initializing sensor positions and an estimating unit for computing state vectors, including 3D position, velocity, acceleration, and quaternion, to accurately model the trainee's motion and compensate for errors.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If depth map-based posture estimation is used, then cost is reduced, but measurement precision deteriorates

Engineering Contradiction:
ImprovecostVSAvoidposture estimation accuracy
Core Design Contradiction:
Ease of manufactureVSMeasurement precision

Solution Approach 1:

The patent combines depth map-based estimation with motion sensor data (acceleration and angular velocity sensors) to create a hybrid estimation system. The converging unit integrates multiple information sources including 3D image information from depth sensors and motion information from motion sensors mounted on body joints, thereby maintaining low cost while improving measurement precision through data fusion.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent introduces a converging unit as an intermediary component that processes and integrates data from multiple sensors. This unit performs initialization by converging motion information and 3D image information, and computes mounting position information to coordinate the different sensor data streams, enabling accurate posture estimation through mediated data integration.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Device complexity

If motion sensors only are used, then device complexity is reduced, but reliability deteriorates due to error accumulation

Engineering Contradiction:
Improvesensor system complexityVSAvoidestimation accuracy over time
Core Design Contradiction:
Device complexityVSReliability

Solution Approach 1:

The patent implements a feedback mechanism where the converging unit continuously integrates motion sensor data with 3D image information from depth sensors. This feedback loop corrects accumulated errors in motion sensor measurements by referencing the more stable depth map data, thereby maintaining reliability over prolonged periods without requiring complex redundant sensor systems.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The converging unit serves as an intermediary that mediates between motion sensor data and depth sensor data. It computes mounting position information and integrates these different data types to produce accurate posture and position estimates, reducing the impact of motion sensor error accumulation while maintaining system simplicity.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Measurement precision

If depth sensor and motion sensors are combined, then measurement precision is improved, but device complexity increases

Engineering Contradiction:
Improveposition and posture estimation accuracyVSAvoidsensor integration complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The converging unit is designed as a multi-functional component that performs multiple tasks: initialization by converging motion information and 3D image information, computation of mounting position information, and continuous posture estimation. This universal component handles diverse sensor data types through a unified processing framework, improving measurement precision while managing device complexity through functional integration.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent transforms different sensor data types into a unified parameter space. The converging unit converts 3D image information and motion sensor readings into consistent mounting position information and state vectors, allowing accurate posture estimation through parameter transformation and integration without requiring complex dedicated processing for each sensor type.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS9378559B2System and method for motion estimation
Publication Date: 2016.06.28 ELECTRONICS & TELECOMM RES INST

AI summary

A system and a method for motion estimation are disclosed. The system for motion estimation in accordance with an embodiment of the present invention includes: a plurality of motion sensors mounted near joints of a body and configured to provide motion information; a depth sensor configured to provide 3-dimensional image information having a 3-dimensional coordinate for each pixel; and a motion estimation device configured to estimate a motion by use of the motion information and the 3-dimensional image information, wherein the motion estimation device includes: a converging unit configured to compute mounting position information of the motion sensors by performing an initialization process by converging the motion information and the 3-dimensional image information; and an estimating unit configured to estimate the motion by computing a state vector including the mounting position information and the motion information.