Drone Localization Using Buffered Images to Correct Drift
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous robots face challenges in maintaining accurate localization within environments with limited processing power and visual features, leading to drift accumulation and reduced accuracy in navigation.
Innovation Solution
The implementation of a buffer image database that stores visual data from previous locations, allowing robots to perform localization when visual features are scarce, by comparing retrieved image data with expected locations to correct their position and overcome drift.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If localization is performed frequently to maintain accurate positioning, then positioning accuracy is improved, but computational processing power and energy consumption increase
Solution Approach 1:
The system performs preliminary actions by storing image data in a buffer database before localization is needed. This allows the robot to have pre-captured images ready for comparison when localization events occur, reducing the computational burden during actual localization operations while maintaining positioning accuracy.
Solution Approach 2:
The system implements periodic localization by triggering localization events based on drift thresholds rather than continuously. The robot monitors its drift accumulation and performs localization only when the drift exceeds a predetermined threshold, reducing computational processing power consumption while maintaining acceptable positioning accuracy.
2Use of energy by moving object
If localization is performed infrequently to save computational processing power, then energy consumption is reduced, but drift accumulation increases and localization accuracy deteriorates
Solution Approach 1:
The system uses feedback by monitoring drift accumulation between localization events. When the drift exceeds a predetermined threshold, the system triggers a localization event to correct the accumulated error. This feedback mechanism ensures that localization accuracy is maintained while minimizing the frequency of localization operations to conserve computational processing power.
Solution Approach 2:
The buffer database stores pre-captured image data from multiple time points, providing ready-to-use reference images for localization. This preliminary action ensures that when localization is triggered, accurate reference data is immediately available, maintaining localization accuracy even when localization events occur infrequently.
3Measurement precision
If the robot waits for favorable visual features to perform localization, then localization accuracy is improved, but time to complete mission increases
Solution Approach 1:
The system performs preliminary actions by continuously capturing and storing image data in the buffer database during normal operation. This ensures that reference images with favorable visual features are already available when localization events are triggered, eliminating the need to wait for favorable conditions and reducing mission completion time while maintaining localization accuracy.
Solution Approach 2:
The system maintains continuity of useful action by continuously populating the buffer database with image data during robot operation. This continuous data collection ensures that reference images are always available, allowing localization to proceed immediately when triggered without interruption or delay for favorable visual conditions.
4Productivity
If the robot performs localization at locations with no visual features, then mission continuity is maintained, but localization accuracy deteriorates
Solution Approach 1:
The buffer database stores reference images captured at various locations and times, including locations with favorable visual features. When the robot reaches a location with no visual features, the system retrieves pre-stored reference images from the buffer for comparison, enabling localization to proceed without delay while maintaining accuracy through the use of high-quality reference data from previous captures.
Solution Approach 2:
The system uses copying by retrieving and comparing stored reference images from the buffer database with current sensor data. This allows the robot to perform localization at locations with no visual features by comparing against copied reference images that contain the necessary visual features, maintaining both mission continuity and localization 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
This approach reduces the frequency of localization computations while maintaining accurate positioning, enabling robots to complete missions by using stored image data to correct for errors and maintain trajectory alignment.
Implementation Method 1
The robot can use visual inertial odometry (VIO) that combines inertial measurements with visual features detected in captured images to determine motion of the robot
Implementation Method 2
The robot can calculate, based on the retrieved image data, an actual location value of the one previous location. The actual location value of the one previous location is compared to an expected location value to determine a location difference
Data Source
AI summary
Methods, systems, and apparatus, including computer programs encoded on computer storage media, for determining that current data captured at a current location of a drone satisfies localization adjustment criteria; in response to determining that the current data captured at the current location of the drone satisfies the localization adjustment criteria, identifying previously captured image data; determining a previous expected location of the drone based on both an expected change in location of the drone and a first previous location determined from other image data captured before the previously captured image data; determining a location difference between the previous expected location of the drone and a second previous location determined from the previously captured image data; and determining the current location of the drone based on the location difference.