Submap Path Planning for Large-Scale Indoor Robot Navigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous navigation methods are inadequate for large-scale indoor environments due to hardware limitations and computational constraints, failing to account for human traffic and dynamic obstacles, and require advanced computational powers and network communication not available in most commercially available mobile robots.
Innovation Solution
A path planning and navigation system for mobile machines that utilizes a map graph divided into submaps, employing a global path planner and local path planner, with collision avoidance and real-time path replanning, using sensors like cameras and Lidar to navigate through large indoor facilities, considering human traffic and dynamic obstacles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If deep learning methods are used for accurate object detection and path planning, then navigation accuracy is improved, but computational power requirements and hardware costs increase significantly
Solution Approach 1:
The patent segments the navigation system into distinct functional modules: map builder, global path planner, and local path planner. Each module performs a specific function with appropriate computational requirements, avoiding the need for deep learning across the entire system. The map builder creates facility maps and identifies navigation features, the global path planner generates overall paths between locations, and the local path planner handles real-time obstacle avoidance, allowing each component to be optimized independently for its specific task.
Solution Approach 2:
The patent replaces deep learning-based perception systems with traditional computer vision and geometric processing methods. Instead of using neural networks for object detection and scene understanding, the system employs algorithmic approaches such as feature extraction, map matching, and path optimization algorithms that are computationally efficient and can run on standard mobile robot hardware without requiring GPU acceleration.
2Reliability
If real-time path replanning is implemented to handle dynamic obstacles, then navigation reliability is improved, but computational time and processing power increase
Solution Approach 1:
The patent performs preliminary path planning at the global level before execution, generating a complete reference path in advance. This global path planner creates a detailed navigation plan that anticipates potential obstacles and planning requirements. When local obstacles are detected during execution, the system only needs to perform local adjustments rather than complete replanning, significantly reducing real-time computational requirements while maintaining reliability.
Solution Approach 2:
The patent implements a dynamic two-layer planning architecture where the global path planner operates at a higher level with less frequent updates, while the local path planner operates continuously at a lower level for immediate obstacle response. This dynamic division allows the system to maintain high reliability through continuous local adjustment without requiring excessive computational resources at real-time speeds, as the computationally intensive global planning is performed less frequently.
3Ease of operation
If map building and path planning are performed on commercially available mobile robots with limited hardware, then ease of operation is improved, but navigation precision and real-time performance deteriorate
Solution Approach 1:
The patent changes the computational parameters and algorithms used in path planning to be suitable for resource-constrained hardware. Instead of using computationally intensive deep learning models, the system employs optimized algorithms with lower computational complexity, such as A* or Dijkstra's algorithm for pathfinding, and efficient map representation structures. These parameter changes enable navigation functionality on commercially available robots with standard processors while maintaining acceptable precision for practical applications.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables efficient, real-time navigation in large indoor facilities with dynamic obstacles, optimizing paths to avoid collisions and adapt to changing conditions, suitable for healthcare environments with emergency protocols.
Implementation Method 1
employing sensors like cameras and Lidar to navigate through large indoor facilities
Implementation Method 2
employing sensors like cameras and Lidar to navigate through large indoor facilities
Data Source
AI summary
Path planning for mobile machine in large scale navigation disclosed. A path for moving a mobile machine is planned by: determining a start map node in a map graph based on a start point in the path and a goal map node in the map graph based on a goal point in the path; determining whether the start map node and the goal map node correspond to the same submap; and if so, planning the path between the start point and the goal point using a real-time path planning method; otherwise, obtaining the path between the start point and the goal point by merging a node path between the start map node and the goal map node, a first real-time path between the start point and a first stop point, and a second real-time path between the goal point and a last stop point.