Occupancy Grid Mapping With Time-Aware Probability Regression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional occupancy mapping technologies fail to consider the dynamic changes in environments, leading to inaccurate occupancy probability assessments due to the equal weighting of recent and outdated sensor observations, especially in dynamic environments.
Innovation Solution
Implementing a time-aware occupancy mapping system that uses Regression to Unknown (RTU) methodology, where each cell's occupancy probability is stored with a timestamp, and the probability is updated based on both the observation and time elapsed, with different regression factors for static and movable objects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If conventional occupancy mapping technologies use equal weighting of sensor observations, then the implementation is simple, but the occupancy probability assessment accuracy deteriorates in dynamic environments
Solution Approach 1:
The patent applies dynamics by making the occupancy probability calculation adaptive to time changes. Instead of using static equal weighting, the system dynamically adjusts the weighting of sensor observations based on the time elapsed since each observation was made. Recent observations are given higher weight while older observations are gradually downweighted, allowing the occupancy map to adapt to environmental changes in dynamic environments.
Solution Approach 2:
The patent changes the parameter of observation weighting from equal to time-dependent. By introducing time as a variable parameter, the system transforms the occupancy probability calculation to account for environmental dynamics. The weighting parameter is adjusted based on time elapsed, enabling the system to differentiate between recent and outdated sensor data.
2Measurement precision
If time-aware occupancy mapping with RTU methodology is implemented, then the occupancy probability assessment accuracy improves, but the system complexity increases
Solution Approach 1:
The patent segments the occupancy probability calculation into distinct components: time-aware weighting of individual sensor observations and Regression to Unknown (RTU) adjustment. By dividing the calculation process into these manageable segments, the system can apply different processing rules to different aspects of the occupancy map, improving accuracy while maintaining organizational clarity in the increased system complexity.
Solution Approach 2:
The RTU methodology implements feedback by continuously adjusting occupancy probabilities based on the elapsed time since last observation. The system feeds back the time information into the probability calculation, creating a closed-loop system that automatically adapts to environmental changes. This feedback mechanism improves accuracy by ensuring outdated information is appropriately downweighted.
3Speed
If equal weighting of sensor observations is used, then the computational processing is fast, but the occupancy map accuracy deteriorates due to outdated information
Solution Approach 1:
The patent applies preliminary action by pre-calculating and storing time-dependent weighting factors or time thresholds. Instead of performing complex calculations for each occupancy probability update, the system prepares time-based parameters in advance, allowing for faster processing during runtime while still achieving time-aware accurate occupancy assessments.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A time-aware occupancy mapping using regression to unknown ( "RTU" ) analysis, and an apparatus to dynamically allocate occupancy probability to a cell in an environment to thereby form a time-aware occupancy map of the environment are disclosed. The apparatus includes a memory circuitry in communication with a processor circuitry, the memory circuitry configured to receive and store probability information from the processor circuitry and to store the probability value and its corresponding time stamp at a probability table. The processor circuitry may be configured to, among others: (1) receive occupancy information, the occupancy information defining whether a first of a plurality of cells (102,104, 106 and 108) in the environment is occupied; (2) determine a first probability value that the first cell is occupied at a first point in time; (3) direct the first probability value and its corresponding timestamp to the memory circuitry to store; (4) determine a second probability value that the first cell is occupied at a second point in time, the second point in time defined by a lapsed interval from the first point in time to the second point in time; and (5) update the memory circuitry to store the second probability value and its corresponding timestamp.