Method for creating grid map of intelligent robot
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing grid map methods for intelligent robots face challenges in balancing precision and storage efficiency, particularly in dense environments, where large grids reduce pathfinding ability but increase storage needs, and small grids enhance resolution but are CPU-intensive.
Innovation Solution
The method employs a grid map with square-shaped grids, where each side is one-third of the robot's diameter, using sensors to update and translate the map in real-time, marking normal passing, obstacle, and edgewise points, and employing a buffer to manage out-of-bounds data efficiently.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If the grid size is made large, then storage of environmental information is reduced and planning time is shortened, but environment resolution is reduced and ability to find paths in dense environment is weakened
Solution Approach 1:
The patent implements dynamic grid translation that adapts to the robot's movement and environmental density. The grid map is not static but translates dynamically based on robot position and detected obstacles, allowing the system to maintain high resolution in relevant areas while optimizing storage usage. This dynamic adaptation resolves the contradiction by making the grid structure flexible rather than fixed.
Solution Approach 2:
The patent changes the parameter of grid resolution dynamically through translation operations. By translating the grid map based on robot movement and environmental features, the system effectively adjusts the resolution parameter in different regions and time steps, allowing high resolution where needed (dense environments) and lower resolution where not needed (open spaces), thus resolving the storage-precision tradeoff.
2Measurement precision
If the grid size is made small, then environment resolution is high and ability to find paths in dense environment is strong, but storage of environmental information is large and CPU intensity increases
Solution Approach 1:
The dynamic grid translation mechanism allows the system to use smaller grids (higher resolution) only when and where necessary - specifically in dense environments where pathfinding requires fine detail. In open areas, the translated grid effectively uses coarser resolution, reducing CPU intensity. This dynamic adaptation to environmental density resolves the contradiction between resolution and computational load.
Solution Approach 2:
The patent applies local quality by translating the grid map to concentrate high-resolution representation in areas of interest (where obstacles and paths are located) rather than uniformly applying high resolution across the entire environment. This allows the system to maintain high environment resolution where needed while minimizing overall storage and CPU requirements.
3Ease of manufacture
If traditional grid methods are used, then implementation is simple, but efficiency is lost in dense environments due to storage-precision tradeoff
Solution Approach 1:
The patent maintains the simplicity of traditional grid methods by using the same basic grid structure and marking mechanisms, but adds dynamic translation capability. This preserves ease of implementation while dramatically improving productivity in dense environments, as the translated grid adapts to environmental density and robot position, providing efficient pathfinding without complex algorithms.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method for creating a grid map of an intelligent robot includes: step (1) controlling a motion of the intelligent robot; step (2) detecting whether an action of the intelligent robot at a current position is an edgewise behavior and detecting whether there is at least one obstacle at the current position, when the action at the current position is the edgewise behavior but there is no obstacle at the current position, proceeding to step (3a), when the action at the current position is the edgewise behavior and there is at least one obstacle at the current position, proceeding to step (3b), when the action at the current position is not the edgewise behavior but there is at least one obstacle, proceeding to step (3c), and when the action at the current position is not the edgewise behavior and there is no obstacle, proceeding to step (3d); step (3a) marking a grid at the current position on the grid map as an edgewise behavior point and returning to step (1); step (3b) marking the grid at the current position on the grid map as the edgewise behavior point, calculating coordinates of a grid where the at least one obstacle is located and marking the grid corresponding to the at least one obstacle on the grid map as at least one obstacle point, and returning to step (1); step (3c) marking the grid at the current position on the grid map as a normal passing point, calculating coordinates of a grid where the at least one obstacle is located and marking the corresponding grid as at least one obstacle point, and returning to step (1); and step (3d) marking the grid at the current position on the grid map as a normal passing point and returning to step (1).