Fusing Inertial and Depth Sensors for Movement Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for body movement recognition using either depth sensors or inertial sensors have limitations when operating under real-world conditions, and there is a lack of development in combining both types of sensors to enhance system robustness.
Innovation Solution
A movement recognition system that combines an inertial sensor and a depth sensor, where the inertial sensor measures the object's inertia and the depth sensor captures three-dimensional shapes using projected light patterns, with a processor determining the type of movement by utilizing a classification model based on signals from both sensors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If only depth sensors are used for body movement recognition, then the system structure is simple, but the recognition robustness is insufficient under real-world conditions
Solution Approach 1:
The patent combines depth sensors and inertial sensors into a unified movement recognition system. The depth sensor captures spatial information while the inertial sensor measures acceleration and orientation, and their data are fused through data association algorithms to achieve more robust and accurate movement recognition than either sensor could provide alone.
2Measurement precision
If only inertial sensors are used for body movement recognition, then the device complexity is low, but the measurement precision deteriorates under real-world conditions
Solution Approach 1:
The patent merges inertial sensor data (acceleration, orientation) with depth sensor data (spatial position, body pose) to achieve more precise movement measurement. The data association algorithm correlates measurements from both sensors, allowing the system to compensate for individual sensor limitations and achieve higher measurement precision.
3Reliability
If both depth sensors and inertial sensors are combined, then the movement recognition robustness is improved, but the device complexity increases
Solution Approach 1:
The patent applies the merging principle by integrating depth sensors and inertial sensors into a coordinated system where both sensor types work together. The data association algorithm fuses the complementary information from both sensors, achieving improved recognition robustness that outweighs the increased device complexity.
Solution Approach 2:
The patent introduces a data association algorithm as an intermediary that processes and correlates data from both depth sensors and inertial sensors. This intermediary component manages the complexity of sensor fusion by providing a structured method for combining measurements, making the integrated system more manageable while maintaining the robustness benefits.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The combined system increases the robustness of movement recognition by compensating for erroneous data from individual sensors, providing more accurate and reliable classification of movements through the use of a classification model that integrates data from both inertial and depth sensors.
Implementation Method 1
The depth sensor is configured to measure a three dimensional shape of the object using projected light patterns and a camera
Data Source
AI summary
A movement recognition system includes an inertial sensor, a depth sensor, and a processor. The inertial sensor is coupled to an object and configured to measure a first unit of inertia of the object. The depth sensor is configured to measure a three dimensional shape of the object using projected light patterns and a camera. The processor is configured to receive a signal representative of the measured first unit of inertia from the inertial sensor and a signal representative of the measured shape from the depth sensor and to determine a type of movement of the object based on the measured first unit of inertia and the measured shape utilizing a classification model.