Occupancy Grid Mapping With Static and Free Space Bayesian Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for characterizing a mobile robot's environment using occupancy grids struggle to efficiently integrate history of measurements, particularly for static obstacles, requiring high-performance computing resources, which is not feasible for embedded microcontroller hardware.
Innovation Solution
A method utilizing binary Bayesian filters to generate static and free space grids, which are then combined using Bayesian fusion, allowing for efficient estimation of cell occupancy by static and free spaces, leveraging integer arithmetic to reduce computational burden.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If standard occupancy grid methods are used to integrate history of measurements, then measurement accuracy is improved, but device complexity increases requiring high-performance computing resources
Solution Approach 1:
The patent changes the computational parameters by using integer arithmetic instead of floating-point operations, and by formulating the Bayesian filter to use only addition and comparison operations. This transforms the computational complexity from requiring high-performance processors to being suitable for embedded microcontrollers, while maintaining occupancy estimation accuracy through the mathematical equivalence of the integer-based Bayesian update rules
Solution Approach 2:
The patent replaces complex probabilistic computations with a simplified integer-based arithmetic system. Instead of using standard floating-point Bayesian calculations, the invention uses integer probabilities and simple arithmetic operations (addition, comparison) to achieve the same occupancy estimation function, thereby reducing computational burden while preserving measurement precision
2Reliability
If standard occupancy grid methods are used to integrate history of measurements, then reliability is improved, but productivity decreases due to computational burden
Solution Approach 1:
The patent modifies the computational parameters by using integer arithmetic with predetermined thresholds instead of continuous floating-point probabilities. This discretization enables real-time processing on embedded devices while maintaining reliable occupancy estimation through the mathematically equivalent integer-based Bayesian update rules that preserve the reliability of occupancy determination
Solution Approach 2:
The patent employs a simplified computational model that uses only the most recent occupancy grid and distance measurements without requiring complex historical data storage or processing. This approach reduces computational burden and enables real-time execution on resource-constrained devices while maintaining sufficient reliability for navigation tasks
Data Source
AI summary
A method for characterizing the environment of a mobile device, wherein, for each iteration at a time t, the following steps are implemented: S10) Acquiring a plurality of distance measurements (zt) in the environment by way of at least one sensor; S20) Generating a pair (wt) of occupancy grids at the time t−1 (OGt-1) and at the time (OGt), each grid (OGt-1, OGt) fusing the distance measurements into a discretized spatial representation of the environment; S30) Generating a static space grid at the time (SGt), Or S40) Generating a free space grid at the time (FGt).