Robot Pose Mapping for Autonomous Cleaning Navigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing robotic navigation systems face challenges in efficiently mapping and navigating complex environments with obstacles, requiring precise path planning and obstacle avoidance while maintaining autonomous operation.
Innovation Solution
The method involves determining a robot's pose using repositionable initialization objects, generating a map of the environment, and projecting training trajectories onto this map to identify unoccupied areas for cleaning, allowing the robot to autonomously navigate and adjust paths to avoid obstacles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the robot uses repositionable initialization objects for pose determination and map generation, then the robot can adapt to environment changes and repositioning without remapping, but the system complexity increases due to object management and association mechanisms
Solution Approach 1:
The patent introduces repositionable initialization objects as intermediaries between the robot and the environment. These objects serve as mediators that carry identification information and establish associations with specific locations, allowing the robot to determine its pose and understand environment changes without requiring complex remapping procedures. The objects act as a simplified interface that bridges the robot's navigation system and the physical environment.
Solution Approach 2:
The initialization objects are pre-configured with identification information and associated with specific locations in the environment before the robot begins operation. This preliminary setup allows the robot to quickly determine its pose and understand environment changes without performing complex analysis during operation. The associations are established in advance, reducing real-time computational requirements.
2Manufacturing precision
If the robot projects training trajectories and sensor data onto a map to identify cleaning areas, then the cleaning precision improves, but the processing time increases due to data projection and expansion operations
Solution Approach 1:
The robot performs training trajectories in advance to collect sensor data and project it onto the map, identifying cleaning areas before actual operation. This preliminary data collection and processing allows the system to pre-determine cleaning paths and areas, reducing the need for real-time processing during actual cleaning operations. The expansion of training trajectories to identify unoccupied grids is performed in advance, enabling faster execution during operational phases.
3Ease of operation
If the robot autonomously navigates using predetermined paths associated with initialization objects, then the ease of operation improves, but the ability to handle dynamic obstacles decreases
Solution Approach 1:
The patent implements a dual-mode navigation system where the robot can operate along predetermined paths associated with initialization objects for routine tasks, but can dynamically switch to reactive obstacle avoidance modes when sensors detect obstacles. The system transitions between pre-planned autonomous navigation and real-time adaptive navigation, allowing the robot to maintain ease of operation for standard tasks while gaining the capability to handle dynamic obstacles when necessary.
Data Source
AI summary
The present disclosure provides systems and methods for robotic navigation, teaching, and mapping. In some cases, the robotic systems and methods may be used to clean an area or environment.