Semantic Obstacle Mapping for Flexible Indoor Navigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional raster map building methods for indoor environments lack reusability, flexibility, and resource efficiency, as they fail to distinguish between different types of obstacles based on their characteristics, leading to inefficient navigation and potential safety hazards.
Innovation Solution
A map building method that classifies obstacles into stabilization, negotiation, risk, and violent-change types based on autonomous movement, interaction, and safety capabilities, allowing for tailored marking, expansion, and updating strategies to enhance navigation flexibility and safety.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional raster map building method is used to represent all environment information on the same map, then the map structure is simple, but the reusability is poor and resource utilization is inefficient
Solution Approach 1:
The patent divides the environment information into multiple map layers: static map layer for fixed obstacles, dynamic obstacle map layer for moving obstacles, and expansion map layer for safety zones. This segmentation allows each layer to be independently managed and reused for specific navigation tasks, improving reusability without requiring a completely complex unified structure.
Solution Approach 2:
The patent introduces a temporal dimension by maintaining historical map versions and allowing navigation to reuse maps from previous time steps when appropriate. This multi-dimensional approach (spatial layers + temporal versions) enhances reusability while managing complexity through organized data structures.
2Reliability
If uniform expansion radius is applied to all obstacle rasters, then the implementation is simple, but the safety is insufficient as it does not consider robot space occupation
Solution Approach 1:
The patent applies different expansion radii to different obstacle rasters based on local conditions. The expansion radius is determined by considering the robot's dimensions and the specific characteristics of each obstacle, creating locally optimized safety zones rather than using a uniform expansion rule throughout the map.
Solution Approach 2:
The expansion radius parameter is dynamically adjusted based on the obstacle type, robot size, and spatial context. This parameter change approach allows the system to maintain safety while adapting to different situations, avoiding the need for overly complex predetermined rules.
3Adaptability or versatility
If all obstacles are treated equally in navigation, then the control logic is simple, but the flexibility is poor and resource utilization is inefficient
Solution Approach 1:
The patent classifies obstacles into different types (static, dynamic, expandable, non-expandable) and applies different navigation strategies to each type. This local differentiation allows flexible navigation planning while keeping the control logic manageable through clear categorization rules.
Solution Approach 2:
The patent implements dynamic obstacle handling where the navigation strategy adapts based on obstacle characteristics and current situation. The system can dynamically adjust which map layers to use and how to interact with different obstacle types, providing flexibility without requiring complex predetermined plans for every scenario.
4Productivity
If hierarchical map management is introduced to improve performance, then the reusability and safety are improved, but the device complexity increases
Solution Approach 1:
The patent implements a hierarchical map structure with distinct layers for different types of information (static, dynamic, expansion zones). This segmentation improves navigation performance by allowing efficient query and update operations on specific layers without processing the entire map, while the hierarchical organization manages the complexity through clear data structures.
Solution Approach 2:
The patent designs the hierarchical map structure to serve multiple functions: navigation planning, obstacle avoidance, path optimization, and historical analysis. This multi-functionality justifies the added complexity by providing comprehensive navigation capabilities from a unified framework, improving overall productivity.
Data Source
AI summary
Provided are a map building method, a navigation method, a device and a system. A detected obstacle is identified, and a type of the obstacle is determined according to an identification result from multiple obstacle types obtained through classification according to an obstacle characteristic; a map is built and the obstacle is marked, and the type of the obstacle is recorded. A newly added obstacle detected on a path is identified during navigation, and obstacle avoidance process is performed according to the type of the newly added obstacle.