Binocular Stereo Camera Grid Map Construction for UAV Navigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current grid map construction methods for navigation, particularly for unmanned aerial vehicles, rely heavily on expensive and limited laser sensors, which are not cost-effective and fail to provide three-dimensional spatial information, leading to two-dimensional maps that do not meet the requirements of vehicles with three-dimensional mobility.
Innovation Solution
A method using a binocular stereo camera to construct a grid map by acquiring disparity images, generating three-dimensional coordinate points, and filtering them to determine grid states based on projection frequency and height, incorporating spatial geometry for adaptive thresholding and visualization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If laser sensors are used for grid map construction, then measurement precision is improved, but device cost and complexity increase
Solution Approach 1:
The patent uses binocular stereo cameras to capture optical images and construct grid maps, replacing expensive laser sensors. The system copies the functional capability of laser sensors (spatial mapping) using visual sensors, achieving comparable measurement precision through image processing and stereo vision algorithms while significantly reducing device cost and complexity
Solution Approach 2:
The patent replaces the mechanical laser ranging system with an optical imaging system. Instead of using laser beams to measure distances directly, the system uses binocular cameras to capture images and derives spatial information through photogrammetry and stereo vision, substituting a mechanical measurement system with an optical one that is cheaper and more versatile
2Device complexity
If two-dimensional grid maps are used, then device complexity is reduced, but loss of information increases due to lack of height data
Solution Approach 1:
The patent transitions from two-dimensional grid maps to three-dimensional grid maps by incorporating height information (Y-axis coordinate) into the map structure. Each grid cell now stores not only horizontal position data but also vertical elevation data, enabling the map to represent terrain height, building structures, and other three-dimensional spatial features while maintaining algorithmic efficiency
Solution Approach 2:
The patent introduces an intermediary processing step that extracts height information from binocular stereo images and integrates it into the grid map structure. The stereo vision system serves as an intermediary that captures depth information and translates it into three-dimensional coordinates, which are then incorporated into the grid map without requiring direct laser measurement
3Device complexity
If binocular stereo cameras are used, then device cost is reduced, but measurement precision may deteriorate compared to laser sensors
Solution Approach 1:
The patent changes the measurement parameters and processing methods to optimize precision for visual sensors. Instead of relying on direct time-of-flight measurements like laser sensors, the system uses feature matching, disparity calculation, and multi-view geometry to derive precise three-dimensional coordinates from images. The grid map construction algorithm is specifically optimized to handle visual data characteristics, improving measurement precision despite using cheaper cameras
Solution Approach 2:
The patent performs preliminary actions to enhance measurement precision before final map construction. This includes calibration of the binocular camera system to determine intrinsic and extrinsic parameters, pre-processing of images to enhance features, and preliminary matching of correspondence points between left and right images. These preliminary steps ensure that the subsequent depth calculation and grid map construction achieve high precision comparable to laser sensors
Data Source
AI summary
The present invention discloses a method for constructing a grid map by using a binocular stereo camera. A high-performance computing platform is constructed by using a binocular camera and a GPU, and a high-performance solving algorithm is constructed to obtain a high-quality grid map containing three-dimensional information. The system in the present invention is easy to construct, so the input data may be collected by using the binocular stereo camera; the program is simple and easy to implement. According to the present invention, the grid height is calculated by using spatial prior information and statistical knowledge, so that a three-dimensional result is more robust; and according to the present invention, the adaptive threshold of grids is solved by using spatial geometry, filtering and screening of the grids are completed, and thus the generalization ability and robustness of the algorithm are improved.