Robot Re-Localization Using Vector Field SLAM After a Pause
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current mobile robot localization systems are costly and complex, requiring extensive mapping and infrastructure modifications, which is not feasible for consumer-grade autonomous navigation, especially in indoor environments like homes, where setup should be simple and affordable.
Innovation Solution
The use of low-cost indoor navigation with active beacons projecting patterns onto the ceiling and leveraging existing infrastructure like WiFi signals, employing Vector Field SLAM to learn signal distributions and estimate pose without prior map knowledge, using low computational resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional mobile robot localization systems are used, then localization accuracy is improved, but system cost and complexity increase significantly
Solution Approach 1:
The patent introduces an intermediary mapping structure (hash map with grid cells) that mediates between sensor measurements and localization computation. This mapping structure simplifies the relationship between sensor data and position estimation, enabling accurate localization with lower computational complexity by avoiding direct complex optimization calculations.
Solution Approach 2:
The patent replaces complex mechanical/computational localization systems (such as laser range finders and extensive mapping infrastructure) with a simplified computational approach using hash maps and grid-based probability distributions. This substitution reduces hardware complexity while maintaining localization accuracy through efficient data structures and algorithms.
2Measurement precision
If extensive mapping and infrastructure modifications are implemented, then localization accuracy is improved, but setup complexity and cost increase
Solution Approach 1:
The system performs self-service by automatically building and updating its environmental map and probability distributions during normal operation. The robot localizes itself without requiring pre-configured infrastructure or manual setup, continuously adapting to the environment through sensor measurements and updating its internal hash map representation automatically.
Solution Approach 2:
The patent performs preliminary actions by pre-initializing the hash map data structure and probability distributions before operation begins. This preliminary setup is minimal and automatic, creating the necessary computational framework without requiring extensive physical infrastructure or manual configuration, enabling the system to start localizing immediately.
3Measurement precision
If high-performance localization systems are used, then localization accuracy is improved, but memory and computational requirements increase
Solution Approach 1:
The patent segments the environment representation into discrete grid cells within hash map buckets, dividing the continuous space into manageable discrete units. This segmentation allows the system to process and store localization data in a distributed, memory-efficient manner, reducing overall computational and memory requirements while maintaining accurate position estimation through localized probability distributions.
Solution Approach 2:
The patent changes the parameter representation from continuous complex probability distributions to discrete grid-based probability values stored in hash maps. This parameter transformation reduces memory requirements by using fixed-size grid cells with simplified probability representations, enabling accurate localization with lower computational energy consumption through efficient data structures.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Vector Field SLAM is a method for localizing a mobile robot in an unknown environment from continuous signals such as WiFi or active beacons. Disclosed is a technique for localizing a robot in relatively large and/or disparate areas. This is achieved by using and managing more signal sources for covering the larger area. One feature analyzes the complexity of Vector Field SLAM with respect to area size and number of signals and then describe an approximation that decouples the localization map in order to keep memory and run-time requirements low. A tracking method for re-localizing the robot in the areas already mapped is also disclosed. This allows to resume the robot after is has been paused or kidnapped, such as picked up and moved by a user. Embodiments of the invention can comprise commercial low-cost products including robots for the autonomous cleaning of floors.