Hybrid Robot Maps for Accurate Localization With Lower Memory
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing robot control systems face challenges in managing and controlling heterogeneous robots efficiently, leading to increased complexity, interference, and reduced autonomous driving performance due to the need for large memory maps and manual command input, especially when dealing with different map formats and types of robots.
Innovation Solution
A hybrid map-based control method that aligns multiple maps in overlapping formats, allowing robots to select the appropriate map format based on location and driving path, and includes a system to detect and prevent interference between robots.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a metric map is used to precisely express the surrounding environment, then the robot can estimate its location accurately, but a large amount of memory is required and semantic information cannot be expressed
Solution Approach 1:
The map is segmented into multiple formats (metric map for precise location estimation and topology map for semantic information). The robot selects and switches between different map formats based on the current situation, rather than using a single comprehensive map that would require large memory
Solution Approach 2:
Different regions of the map have different qualities and formats. The metric map provides detailed spatial information where needed, while the topology map provides semantic information in other regions. The robot uses the appropriate map format locally based on its current location and task requirements
2Quantity of substance
If a topology map is used to express the environment in a simple form, then memory usage is reduced and path planning is simplified, but the robot cannot estimate its location accurately using map information alone
Solution Approach 1:
The system merges multiple map formats (metric map and topology map) into a hybrid map structure. This allows the robot to benefit from both the memory efficiency of topology maps and the location estimation accuracy of metric maps by switching between them as needed
3Productivity
If multiple heterogeneous robots are controlled individually, then each robot can perform its assigned task, but interference between robots occurs and manual control is required
Solution Approach 1:
A universal robot management system is introduced that can manage multiple heterogeneous robots through a common interface and coordinate their actions. This system detects potential interferences and grants driving priority to appropriate robots, allowing individual robots to perform tasks autonomously while preventing conflicts
4Ease of operation
If commands are issued manually to robots, then control is straightforward, but the management process becomes complex and time-consuming as the number of robots increases
Solution Approach 1:
The robot management system operates autonomously to detect interference situations and grant driving priority to appropriate robots without requiring manual intervention. This self-service capability reduces the time and complexity of managing multiple robots while maintaining straightforward control operations
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A control method of a hybrid map-based robot includes receiving satellite information of a target region, generating a first map of the target region based on the satellite information, dividing a region in the first map according to a preset reference, receiving 3D information (point cloud data) for a region including a road available for the robot to drive on among the divided regions, generating a second map based on the 3D information, generating a hybrid map by aligning the first map and the second map in an overlapping manner, and determining a location of the robot within the target region, wherein the robot drives using one of the different formats of maps forming the hybrid map depending on a location within the target region.