Position Estimation Apparatus for Autonomous Parking Route Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing automatic parking systems face challenges in accurately estimating the position of a mobile object moving autonomously along a complex route, leading to position estimation errors and difficulties in accurate autonomous movement.
Innovation Solution
A position estimation apparatus that includes an image information calculation unit for extracting and associating image feature points, a travel amount calculation unit for estimating the mobile object's position, a three-dimensional image feature point calculation unit for estimating the object's position in real space, a position correction unit for reducing position errors, and a route regeneration judgment unit for determining if the route needs to be adjusted based on the corrected position.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the automobile moves autonomously along a complicated route with multiple steering wheel turns, then the parking maneuver can be completed, but position estimation errors accumulate and autonomous movement accuracy deteriorates
Solution Approach 1:
The patent implements feedback by calculating the position deviation between the actual automobile position and the target position on the route, then regenerating the route based on this deviation. The position-in-route judgement unit continuously monitors whether the automobile has reached preset spots, and the route regeneration judgment unit uses this feedback to determine if route regeneration is needed, thereby correcting accumulated position estimation errors.
Solution Approach 2:
The patent applies preliminary action by presetting multiple spots along the route in advance and preparing route regeneration strategies before position errors become critical. The system pre-calculates alternative routes and regeneration parameters, enabling quick correction when deviation is detected without waiting for complete position failure.
2Extent of automation
If position estimation is performed continuously along the route, then autonomous movement can be maintained, but position errors accumulate over time and distance
Solution Approach 1:
The system continuously monitors position reliability by calculating deviation between estimated position and actual position relative to preset route spots. When deviation exceeds thresholds, the feedback mechanism triggers route regeneration to reset position accuracy, maintaining reliable autonomous movement over extended periods.
Solution Approach 2:
The patent makes the position estimation system dynamic by adaptively switching between continuous estimation mode and route regeneration mode based on real-time position accuracy. The system adjusts its operation dynamically, regenerating routes when needed and continuing normal estimation when reliable, rather than using a fixed approach.
3Measurement precision
If the route is regenerated frequently to correct position errors, then position accuracy is maintained, but system complexity and processing load increase
Solution Approach 1:
The patent uses parameter changes by setting specific threshold values for position deviation that trigger route regeneration. Instead of continuously regenerating routes, the system monitors position parameters and only regenerates when deviation exceeds predefined thresholds, balancing accuracy maintenance with system simplicity.
Solution Approach 2:
The system applies local quality by implementing route regeneration only in specific locations where position deviation is detected, rather than globally regenerating the entire route continuously. The position-in-route judgement unit identifies local spots along the route where correction is needed, and route regeneration is performed selectively at those locations.
Data Source
AI summary
A position estimation apparatus includes: an image information calculation unit that extracts each of image feature points from a plurality of images which are sequentially captured and associates the image feature points with each other between the plurality of images; a travel amount calculation unit that estimates the position of the mobile object; a three-dimensional image feature point calculation unit that estimates a three-dimensional position of the image feature point in a real space on the basis of a result of associating the image feature points with each other and the position of the mobile object; a position correction unit that corrects the three-dimensional position and the position of the mobile object so as to reduce a position error between a position of the three-dimensional position, when the three-dimensional position is reprojected onto any one of the plurality of images, in the relevant image and the image feature point; a route regeneration judgment unit that judges whether or not to regenerate the route on the basis of a result of comparison between the corrected position of the mobile object and the route.