Contextual Keepout Maps for Mobile Robot Navigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing map technologies for robots lack contextual information, making it difficult for them to navigate and perform tasks effectively in dynamic environments, as they rely on outdated blueprints and lack understanding of keepout regions and object locations.
Innovation Solution
The implementation of contextual map layers, which combine localization maps with additional data layers providing semantic information about the environment, such as floor types, obstacles, and keepout regions, allowing robots to understand their surroundings and adjust their navigation and tasks accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If SLAM algorithms are used to generate maps, then navigation capability is improved, but map accuracy deteriorates due to drift accumulation
Solution Approach 1:
The patent merges SLAM-generated maps with pre-existing map data (from GPS, aerial imagery, or other sources) to create a hybrid map system. This combination allows the robot to benefit from the local detail of SLAM while using the pre-existing map to correct drift accumulation and maintain global accuracy, thereby resolving the contradiction between navigation capability and map accuracy.
Solution Approach 2:
The patent introduces a map matching module as an intermediary that compares SLAM-generated features with pre-existing map features. This mediator identifies and corrects drift by finding correspondences between the two map sources, allowing the system to maintain both the real-time navigation benefits of SLAM and the accuracy of pre-existing maps.
2Adaptability or versatility
If robot explores unknown environments to build maps, then adaptability is improved, but time consumption increases
Solution Approach 1:
The patent uses pre-existing maps (aerial imagery, GPS data, or previously collected map data) to provide advance information about the environment. This preliminary action allows the robot to navigate more efficiently by leveraging pre-collected spatial information, reducing the time needed to explore and map unknown areas while maintaining adaptability to new environments.
Solution Approach 2:
The patent segments the mapping process into areas covered by pre-existing maps and areas requiring new exploration. By identifying regions where pre-existing map data is available, the robot can skip redundant exploration in those areas and focus only on truly unknown regions, thereby reducing overall time consumption while maintaining environmental adaptability.
3Adaptability or versatility
If robot operates in dynamic environments, then versatility is improved, but localization accuracy deteriorates due to changing features
Solution Approach 1:
The patent employs dynamic map updates where the map is continuously refined as the robot operates in the environment. By dynamically adjusting the map based on new sensor data and detecting changes in the environment, the system maintains localization accuracy even as environmental features change, allowing operation in dynamic environments without sacrificing precision.
Solution Approach 2:
The patent implements feedback mechanisms where the robot continuously compares its observed environment with the map data and adjusts its localization accordingly. This feedback loop allows the system to detect and adapt to environmental changes, maintaining accurate localization in dynamic environments by constantly refining its understanding of the current state versus the stored map.
Data Source
Figure 1
Figure 2
Figure 3A~3B
AI summary
A mobile robot can navigate within an area using a keepout map. The keepout map identifies portions of the area through which the robot can safely navigate, and likewise identifies portions of the area through which the robot is unable to navigate. The keepout map can be generated based on external data (such as blueprints or local area maps) or based on data from sensors included on or within the robot (such as cameras or LIDAR sensors). The keepout map can also be updated as the robot navigates within the area, for instance to identify new obstacles within the area. The robot navigates through the area by selecting a route that avoids the portions of the area that the robot is unable to navigate identified by the keepout map. The updated keepout map can be provided to a central server, for storage or distribution to one or more other robots.