Mobile Pose Estimation Using DVS and Radar Depth Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current pose estimation methods using RGB and IMU sensors suffer from low accuracy in complex lighting conditions, such as excessively strong or dim light environments.
Innovation Solution
Integrate a dynamic vision sensor (DVS) and a single-line lidar to enhance feature extraction and fusion, leveraging the DVS's ability to capture dynamic changes and the lidar's robustness to lighting conditions, with feature enhancement and alignment to improve pose estimation accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If RGB color sensor and IMU are used for pose estimation, then the system structure is simple, but pose estimation accuracy deteriorates in complex lighting conditions
Solution Approach 1:
The patent combines DVS (dynamic vision sensor) and single-line lidar into a unified pose estimation system. The DVS captures dynamic visual information while the lidar provides depth data independent of lighting conditions. By merging these two sensors and their respective data streams, the system achieves robust pose estimation that overcomes the limitations of individual sensors in complex lighting environments.
2Measurement precision
If DVS and single-line lidar are integrated for pose estimation, then pose estimation accuracy in complex environments is improved, but device complexity increases
Solution Approach 1:
The patent segments the pose estimation process into distinct functional modules: DVS data processing, lidar data processing, feature enhancement, and pose calculation. Each module handles specific aspects of the data stream independently. This segmentation allows the complex multi-sensor system to be managed through modular processing steps, reducing the practical complexity despite the advanced functionality.
3Measurement precision
If feature enhancement is performed on depth information, then pose estimation accuracy is improved, but processing time increases
Solution Approach 1:
The patent performs feature enhancement on depth information as a preliminary step before final pose estimation. By pre-processing the depth data to enhance features (such as edges, corners, and significant depth variations), the system prepares optimized input for the pose calculation algorithm. This preliminary action reduces the computational burden during the actual pose estimation, effectively managing processing time while maintaining high accuracy.
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
Enhances pose estimation accuracy in complex environments by complementing DVS and lidar data, ensuring accurate determination of mobile apparatus pose regardless of lighting conditions, while also reducing power consumption and costs.
Implementation Method 1
obtaining first sensing data, where the first sensing data is obtained by using a dynamic vision sensor (dynamic vision sensor, DVS) of the mobile apparatus by detecting physical space in which the mobile apparatus is located
Implementation Method 2
the second sensing data is obtained by using a single-line lidar of the mobile apparatus by detecting the physical space in which the mobile apparatus is located
Data Source
Figure 1
Figure 2~3
Figure 4~5
AI summary
A pose estimation method and a related apparatus are provided. The method includes: obtaining first sensing data and second sensing data, where the first sensing data is obtained by using a DVS of a mobile apparatus by detecting physical space in which the mobile apparatus is located, and the second sensing data is obtained by using a 2D radar of the mobile apparatus by detecting the physical space in which the mobile apparatus is located; performing feature enhancement on first depth information based on the first sensing data, to obtain second depth information, where the first depth information is obtained by fusing the first sensing data and the second sensing data; and determining a pose of the mobile apparatus based on the second depth information. A characteristic that the radar is not affected by intensity of light in an environment and a characteristic that the DVS is good at capturing information about a dynamic change in the environment are used, and fusion and feature enhancement are performed on the sensing data obtained by using the DVS and the 2D radar, so that in a complex environment, pose estimation accuracy can be improved.