Robot SLAM Mapping with Overlap Stitching for Faster Coverage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous robots face challenges in efficiently mapping and navigating environments without redundant coverage, as traditional SLAM techniques require extensive time and perimeter tracing, which is unsuitable for service-oriented tasks like robotic vacuums.
Innovation Solution
The implementation of a Light Weight Real Time SLAM Navigational Stack that reduces computational burden, allowing for faster mapping and navigation, enabling robots to generate complete maps in minutes with minimal coverage, and autonomously learn calibration of sensors, while using a tiered sensing system and Hebbian-based correlations between sensors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional SLAM techniques are used for mapping and navigation, then mapping accuracy and completeness are improved, but time consumption and computational burden increase significantly
Solution Approach 1:
The patent segments the mapping process into two distinct phases: a fast initial mapping phase that quickly establishes the basic environment structure, and a subsequent refinement phase that improves accuracy in specific areas. This segmentation allows the robot to achieve functional mapping results rapidly while reserving computational resources for targeted accuracy improvements, directly resolving the contradiction between mapping accuracy and time consumption.
Solution Approach 2:
The patent implements partial action by performing mapping refinement only in areas that require higher precision, rather than uniformly refining the entire map. The system identifies regions needing improvement and applies computational resources selectively, achieving adequate mapping accuracy for service tasks without the excessive time consumption of complete map refinement.
2Measurement precision
If traditional SLAM techniques with extensive perimeter tracing are used, then complete environment coverage is achieved, but productivity and task completion speed decrease
Solution Approach 1:
The patent applies partial action by achieving sufficient coverage for service tasks without performing exhaustive perimeter tracing. The system determines when adequate coverage has been reached based on task requirements, allowing the robot to transition from mapping to service tasks earlier, thereby maintaining productivity while achieving necessary coverage completeness.
Solution Approach 2:
The patent segments the navigation task into coverage acquisition phases and service task phases. Rather than completing exhaustive perimeter tracing before beginning service tasks, the robot performs sufficient coverage to enable service operations, then divides remaining coverage tasks into separate segments that can be performed during service operations or deferred, significantly improving overall productivity.
3Measurement precision
If extensive sensor data collection and processing is performed, then mapping accuracy is improved, but energy consumption increases
Solution Approach 1:
The patent segments data collection and processing into essential measurements for basic mapping and optional refined measurements for accuracy improvement. The system performs only the essential measurements needed for functional mapping during normal operation, reserving energy-intensive refined measurements for specific situations, thereby maintaining mapping accuracy while significantly reducing battery consumption.
Solution Approach 2:
The patent dynamically changes processing parameters based on operational context, adjusting the level of measurement precision and processing intensity according to task requirements and battery status. This allows the system to achieve adequate mapping accuracy with reduced energy expenditure by lowering measurement and processing parameters when full precision is not required.
Data Source
AI summary
A method including detecting an object in a line of sight of at least one sensor; adjusting a current path of the robot to include a detour path around the object, instructing the robot to resume along the current path after avoiding the object, discarding at least some data collected by sensors of the robot in overlapping areas covered, inferring previously visited areas and unvisited areas, generating a planar representation of a workspace of the robot by stitching data collected by at least some sensors of the robot at overlapping points, and presenting at least the planar representation and coverage statistics on an application of a communication device.