Mobile Robot Map Production Using Hybrid Sensor Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional SLAM systems face challenges in generating map information with high accuracy, especially when environmental changes occur, as they rely heavily on internal sensors and may struggle to update map information in real-time, leading to discrepancies between actual maps and map information.
Innovation Solution
The proposed method involves a mobile robot equipped with internal and external sensors, which acquires information on movement and environmental changes, updates map information by adjusting variance values based on sensor overlap and environmental data, and uses environmental sensors to enhance map reliability and accuracy, even in areas outside direct detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If SLAM systems rely heavily on internal sensors for map generation, then the system complexity is reduced, but the map accuracy and reliability deteriorate when environmental changes occur
Solution Approach 1:
The patent combines internal sensors (odometry) with external sensors (laser range finders, cameras) to create a hybrid SLAM system. The external sensors provide accurate environmental measurements while internal sensors provide continuous position tracking, resolving the contradiction between system complexity and map accuracy by merging complementary sensing modalities.
Solution Approach 2:
The patent introduces environmental sensors as intermediary components that mediate between the mobile robot and the surrounding environment. These sensors act as intermediaries to capture environmental changes and feed this information back to the map updating module, enabling accurate map generation without directly increasing internal sensor complexity.
2Device complexity
If SLAM systems use only internal sensors for position estimation, then the device complexity is lower, but the reliability of map information deteriorates when environmental changes occur
Solution Approach 1:
The patent implements feedback mechanisms where environmental sensors continuously monitor the surrounding environment and feed this information back to the map updating module. This feedback loop enables the system to detect environmental changes and adjust map information accordingly, significantly improving map reliability without requiring complex internal sensor systems.
Solution Approach 2:
The patent makes the map updating process dynamic by continuously adjusting map information based on real-time environmental sensor data. The system transitions from static map generation to dynamic map updating, where map reliability is continuously maintained through adaptive adjustments based on environmental changes detected by external sensors.
3Speed
If SLAM systems do not update map information in real-time, then the processing speed is higher, but the accuracy of map information deteriorates when environmental changes occur
Solution Approach 1:
The patent implements periodic map updating where the map information is updated at regular intervals based on environmental sensor data. This periodic action balances processing speed with accuracy by updating maps frequently enough to capture environmental changes while maintaining efficient processing throughput, resolving the contradiction between speed and accuracy.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
There is provided a map production method to be performed by a mobile robot which moves in a first region and includes a first sensor and a second sensor. The map production method includes acquiring first information from the first sensor, acquiring second information from the second sensor, acquiring third information from a third sensor provided in the first region, acquiring fourth information indicating a detection region of the second sensor calculated from the first information and the second information, acquiring fifth information indicating a detection region of the third sensor calculated from the third information, and updating map information of the first region for a third region, including a first object, if a the second region overlaps with the third region is judged to be present from the fourth information and the fifth information.