Mobile Vehicle Mapping That Filters Moving Objects
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing localization and mapping algorithms for mobile machines in dynamic environments fail to effectively map fixed objects that may be moved later, leading to ineffective map recognition and reconstruction upon subsequent passage through the same region.
Innovation Solution
A method and system that utilize on-board sensors, such as lidar and image sensors, to differentiate between fixed and mobile objects, constructing a map based on fixed objects and filtering out data from mobile objects, allowing for robust and reusable mapping by focusing on persistent environmental features.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If the localization and mapping algorithm maps all detected objects including moving objects, then the map construction is simpler, but the map becomes less robust and cannot be easily reused in subsequent passages
Solution Approach 1:
The patent segments the environment into fixed objects and moving objects. The localization and mapping algorithm processes these two categories differently: fixed objects are mapped for long-term reuse, while moving objects are tracked separately. This segmentation allows the system to maintain a robust, reusable map of the static environment while still accounting for dynamic elements, resolving the contradiction between map construction simplicity and map reusability.
2Reliability
If the algorithm filters out moving objects from the map, then the map is more robust and reusable, but the algorithm complexity increases
Solution Approach 1:
The patent introduces dynamic object tracking alongside static mapping. The system maintains two parallel processes: a standard SLAM algorithm for fixed objects and a moving object tracking mechanism that identifies and excludes moving objects from the permanent map. This dynamic approach increases algorithm complexity but enables robust, reusable maps by filtering out transient elements while preserving the static environment structure.
3Adaptability or versatility
If the algorithm restarts map construction during subsequent passages, then it adapts to current objects, but productivity and efficiency are reduced
Solution Approach 1:
The patent performs preliminary action by constructing a reusable map of fixed objects during the first passage and storing it for future use. During subsequent passages, the system loads and reuses this pre-constructed map instead of restarting the mapping process. This preliminary map construction enables the system to quickly adapt to the environment by overlaying current sensor data on the existing map, significantly improving mapping efficiency while maintaining adaptability to the static environment structure.
Data Source
Figure 1~3
Figure 4~6
AI summary
The invention relates to a localisation and mapping method used by a mobile vehicle in an environment, said method comprising the following steps: the determination of the type of an object located in an area of the environment, on the basis of the data received from an on-board sensor in the mobile vehicle; and implementation of a localisation algorithm using detection data, without taking into account the detection data relating to said area or said object when the determined type is a type of mobile object.