Depth-Camera Robot Navigation Using Floor Height 3D Points
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current robots using SLAM technology face challenges with position estimation due to high computational load, image blurring at high speeds, short lifespan and high cost of LIDAR sensors, and difficulty in sensing environments at varying heights.
Innovation Solution
A robot equipped with a depth camera and processor that generates 3D points from depth images, identifies points of specific heights to determine the floor, and controls movement based on these points, allowing for efficient mapping and navigation without the limitations of traditional sensors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If a vision sensor such as a camera is used for position estimation, then cost is reduced and miniaturization is possible, but computational load increases and image blurring occurs at high speeds
Solution Approach 1:
The patent extracts only the essential depth information needed for position estimation from the full image data, processing a simplified representation rather than complete high-resolution images. This reduces computational load while maintaining the cost and size advantages of camera-based sensors.
Solution Approach 2:
The system performs preliminary processing to identify and extract depth information before full image processing is required. By pre-processing the depth data and extracting key features in advance, the computational burden during actual navigation and position estimation is significantly reduced.
2Measurement precision
If a LIDAR sensor is used for position estimation, then measurement precision is improved, but cost increases, lifespan decreases, and miniaturization becomes difficult
Solution Approach 1:
The patent creates a digital 3D model (copy) of the environment using depth information from the camera, which then serves as the basis for position estimation. This virtual representation replicates the functionality of LIDAR point clouds while using the more affordable camera hardware, achieving similar measurement precision without the high cost and size constraints of LIDAR systems.
Solution Approach 2:
The patent replaces the mechanical LIDAR sensing system with an optical camera-based depth sensing system. By substituting the active illumination and mechanical scanning approach of LIDAR with passive optical depth capture, the system achieves comparable precision while enabling miniaturization and reducing cost.
3Ease of manufacture
If a 2D LIDAR is used, then cost is reduced compared to 3D LIDAR, but sensing capability is limited to fixed height only
Solution Approach 1:
The patent transitions from 2D depth mapping to 3D spatial understanding by constructing a three-dimensional model of the environment. This dimensional enhancement allows the system to capture and process depth information at multiple heights simultaneously, providing versatile sensing capability across various elevations while maintaining the cost advantages of camera-based sensors over 3D LIDAR.
Data Source
AI summary
Provided in the present disclosure are a robot and a control method therefor. The robot includes: a depth camera; a driver; and a processor for acquiring a depth image by performing photographing through the depth camera, generating a plurality of 3D points on a three-dimensional (3D) space corresponding to a plurality of pixels, based on depth information about the plurality of pixels of the depth image, identifying a plurality of 3D points having a preset height value, based on a driving bottom surface of the robot in the 3D space from among the plurality of 3D points, and controlling the driver to move the robot based on the identified plurality of 3D points.