Autonomous Mobile System Localization Using Segmented Object Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional autonomous mobile systems face localization errors due to the presence of moving objects in the environment, as they mismatch shape data from moving objects with pre-existing maps, leading to loss of target path and control.
Innovation Solution
An autonomous mobile system that divides space into segments, uses sensors to differentiate between stationary and moving objects based on detection frequency, and localizes the mobile body by matching stationary objects to pre-defined map data, ensuring accurate navigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If shape data of peripheral objects is matched to the map for localization, then the mobile body can navigate autonomously, but localization error increases when moving objects are present
Solution Approach 1:
The patent segments the localization process by dividing detection objects into two categories: stationary objects (buildings, trees, etc.) and moving objects (vehicles, pedestrians, etc.). The system separately processes shape data for stationary objects when matching with the map, while excluding or differently handling moving objects. This segmentation prevents moving objects from causing localization errors while still allowing autonomous navigation functionality.
2Quantity of substance
If all detected objects are used for map matching, then more data is available for localization, but mismatching occurs with moving objects
Solution Approach 1:
The patent extracts and identifies moving objects from the set of all detected peripheral objects using recognition means. Once moving objects are identified, they are separated out from the shape data used for map matching. This extraction process ensures that only stationary objects are used for matching with the pre-stored map, preventing mismatching errors while still utilizing available shape data from the environment.
Data Source
AI summary
An autonomous mobile system for a mobile body which moves while localizing itself in a space includes a shape detector and travel distance detector for measuring whether objects are present in set regions determined by dividing a three-dimensional space into a plurality of segments, a storage device into which map data is stored that indicates a region of the set regions that has been set as having a stationary object in the region, a determining section that determines, from frequency of the object detection by the shape detector during a predetermined time for each of the set regions, whether the object that has been detected in each set region is a stationary object or a moving object, and a localizer that localizes a vehicle by matching the region that the determining section has determined to have a stationary object in the region, and the map data.