Hexagonal Map Path Planning Using Context and Spatial Similarity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional AI technologies for robot navigation require large volumes of training data, are complex, time-consuming, and energy-intensive, and struggle with dynamic environments and hexagonal grid representations, which are more efficient like the human brain's spatial mapping.
Innovation Solution
A processor-implemented method for context-based path planning in hexagonal spatial maps using RGB values to identify objects, create object and context databases, and calculate a Spatial Similarity Quotient (SSQ) for dynamic path planning, mimicking the human brain's efficient spatial information handling.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional AI techniques (machine learning, deep learning) are used for robot navigation, then the system can learn from training data, but it requires huge volume of training data, is very complex, time consuming, and consumes a lot of energy
Solution Approach 1:
The patent copies the human brain's spatial mapping mechanism (entorhinal-hippocampal system) to create an artificial spatial mapping system. Instead of using conventional AI that learns from huge datasets, the system implements a biologically-inspired spatial index that directly maps environmental features to navigation decisions, dramatically reducing computational requirements and energy consumption while maintaining adaptability
Solution Approach 2:
The patent replaces the mechanical/computational system of conventional AI training with a biologically-inspired neural mechanism. The artificial spatial mapping system uses simplified neural computations that mimic brain function, substituting the heavy mechanical processing of traditional machine learning with more efficient bio-inspired algorithms
2Adaptability or versatility
If conventional AI techniques are used for robot navigation, then the system can make decisions, but it needs to be re-trained every time features change, making it complex and time consuming
Solution Approach 1:
The patent implements a dynamic spatial mapping system that continuously updates and adapts to new environmental features in real-time. The artificial spatial index dynamically adjusts to new objects and layouts without requiring re-training, allowing the robot to navigate novel environments immediately while maintaining decision-making capabilities
Solution Approach 2:
The system pre-establishes the spatial mapping framework and neural mechanisms before encountering specific environments. This preliminary setup allows the robot to immediately begin navigating new spaces using its pre-wired spatial understanding, eliminating the time required for re-training while maintaining adaptability to different features
3Ease of manufacture
If square grid maps are used for robot navigation, then the environment can be represented, but hexagonal grid maps are more efficient like the human brain's spatial mapping
Solution Approach 1:
The patent transitions from the symmetric square grid to an asymmetric hexagonal grid structure. The hexagonal grid better mimics the human brain's spatial representation and provides more efficient path planning by reducing the number of steps required to reach goals while maintaining ease of environment representation through the tessellating hexagonal cells
4Use of energy by moving object
If context-based path planning with spatial similarity comparison is implemented, then computational effort and energy consumption are reduced, but the system must process and compare multiple transformed map variations
Solution Approach 1:
The patent pre-stores multiple transformed versions of reference maps (rotated, flipped, scaled) in advance. When navigation is needed, the system simply compares the current spatial map against these pre-prepared variations using the spatial similarity quotient, avoiding the computational burden of real-time transformations while reducing energy consumption
Data Source
Figure 1
Figure 2
Figure 3A
AI summary
Path planning for a robot is a compute intensive task. For a dynamic environment this is more cumbersome where position and orientation of objects changes often. Embodiments of the present disclosure provide systems and methods for context based path planning for vector navigation in hexagonal spatial maps. A 2-D environment is represented into a hexagonal grid map that includes hexagonal grid cells, objects are identified based on a comparison of RGB value associated with contiguous cells. Candidate contexts are determined based on objects identified. The hexagonal grid map is rotated at various angles and compared with pre-defined map(s) to determine quantitative measure of similarity for contexts identification from the candidate contexts, based upon which a path is dynamically planned for easy and efficient vector navigation within the hexagonal grid map. The embodiments further enable generating paths for different contexts using navigable common object(s) identified between intersections of the different contexts.