Positioning System for Automatic Carriers Using Sensor Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional positioning techniques for automatic carriers, relying on odometers and direction sensors, face errors when navigating bumpy roads, and require frequent recalibration of environment video cameras, which can be inefficient and fail when the carrier is obstructed.
Innovation Solution
A positioning system that dynamically updates pose information using both a first positioning device (odometer and direction sensor) and a second positioning device (video camera) by generating candidacy pose information within an error range, transforming it between coordinate systems, and selecting the pose with the smallest error to correct parameters, allowing for continuous localization without prior calibration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If odometer and direction sensor are used for positioning, then positioning function is provided, but positioning accuracy deteriorates on bumpy roads and over time
Solution Approach 1:
The patent combines multiple positioning devices (odometer, direction sensor, and video camera) into a unified positioning system. The operation module integrates data from all devices to provide both short-term positioning (from odometer/direction sensor) and long-term positioning (from video camera), resolving the contradiction between reliability of continuous positioning and accuracy over time.
Solution Approach 2:
The operation module serves as an intermediary that processes and coordinates information between different positioning devices. It generates candidate pose information from one device and validates it against another device, acting as a mediator that reconciles the differing accuracy characteristics of each positioning method.
2Measurement precision
If environment video camera is used for localization, then positioning accuracy is improved, but calibration frequency increases and operation efficiency decreases
Solution Approach 1:
The system performs preliminary calibration of the video camera only once during system setup. After calibration, the operation module automatically uses the calibrated parameters to transform image coordinates to world coordinates without requiring repeated manual calibration, thus maintaining accuracy while improving operational efficiency.
Solution Approach 2:
The operation module automatically performs the localization function by processing image data through the pre-calibrated video camera parameters. The system serves itself by automatically generating candidate pose information and validating it against sensor data without requiring manual intervention for each positioning operation.
3Measurement precision
If only environment video camera is used for positioning, then localization is provided, but positioning fails when carrier is hidden by environment objects
Solution Approach 1:
The patent segments the positioning function into two independent parts: short-term positioning using odometer and direction sensor, and long-term positioning using video camera. Each segment operates independently and can compensate for the other's failures, ensuring reliable positioning whether the carrier is visible or hidden.
Solution Approach 2:
The system changes the operational parameters by switching between different positioning devices based on environmental conditions. When the carrier is hidden, the system relies on sensor data parameters; when visible, it uses video camera parameters, dynamically adapting to maintain reliability.
Data Source
AI summary
A positioning system and a method thereof are provided. In the positioning method, a first and a second pose information of a moving device are obtained by a first positioning device and a second positioning device respectively, wherein the first pose information corresponds to the second pose information. In addition, a plurality of first candidacy pose information is generated in an error range of the first pose information. Furthermore, a plurality of second candidacy pose information is generated according to the first pose information respectively. One of the second candidacy pose information having a smallest error derived from the second pose information is selected for updating the pose information of the first positioning device and parameter information of the second positioning device. Thereby, pose information of the moving device is updated and parameter information of the second orientation devices is calibrated simultaneously.