Temporal Probability Updates in Autonomous Vehicle Occupancy Grids
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous vehicle control systems face challenges in balancing the influence of new and outdated sensor measurement data, leading to potentially hazardous decision-making in incompletely known environments, while requiring modest computational resources.
Innovation Solution
A method utilizing a probabilistic occupancy grid that associates each cell with multiple obstacle hypotheses and probability scores, updating these scores based on new measurement data and evolving them over time to account for data aging, with computational efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the occupancy grid continuously updates probability scores for all cells based on new sensor data, then the reliability of vehicle control decisions is improved, but the computational resources required increase significantly
Solution Approach 1:
The patent applies local quality by differentiating between corresponding cells (those intersected by sensor rays) and non-corresponding cells. Only corresponding cells undergo full probability score updates based on new sensor measurements, while non-corresponding cells have their scores evolved using temporal decay functions. This localized processing approach maintains reliability for observed areas while significantly reducing computational resources compared to updating all cells uniformly.
Solution Approach 2:
The patent segments the occupancy grid cells into two distinct categories: corresponding cells and non-corresponding cells. This segmentation allows the system to apply different processing strategies to different cell types, updating only the necessary portion of the grid with new sensor data while efficiently managing computational resources through selective processing.
2Use of energy by moving object
If the system uses outdated measurement data for vehicle control decisions, then computational resources are conserved, but the reliability and safety of control decisions deteriorate
Solution Approach 1:
The patent implements dynamics by continuously evolving probability scores over time through temporal decay functions. For non-corresponding cells without recent sensor updates, the system dynamically adjusts probability scores based on elapsed time, preventing stale data from dominating decisions while avoiding the computational cost of frequent full-grid updates. This dynamic adaptation maintains safety by ensuring outdated information naturally loses influence.
Solution Approach 2:
The system incorporates feedback mechanisms where probability scores are continuously refined based on new sensor measurements and temporal evolution. The feedback loop ensures that control decisions are based on the most current reliable data available, with the system automatically adjusting the weight of historical data versus new observations to maintain safety without excessive computational burden.
3Device complexity
If the occupancy grid uses a single obstacle hypothesis per cell, then the device complexity is reduced, but the adaptability to handle multiple possible obstacle scenarios deteriorates
Solution Approach 1:
The patent applies parameter changes by representing multiple obstacle hypotheses within each grid cell through different probability score parameters. Instead of maintaining separate complex data structures for each hypothesis, the system uses probabilistic parameters to encode multiple possible states (obstacle present, obstacle absent, uncertain). This approach maintains relatively simple grid structure while achieving high adaptability in handling multiple obstacle scenarios through parameter-based representation.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method of controlling an autonomous vehicle which is movable on a surface, comprising: initializing an occupancy grid (230) with reference map data, each cell (231) in the occupancy grid being associated with a plurality of obstacle hypotheses and respective probability scores; repeatedly obtaining measurement data from sensors (250); associating the obtained measurement data with corresponding cells (231-5); in each of said corresponding cells, updating the probability score of at least one obstacle hypothesis in accordance with the obtained measurement data; outside (231-1, 231-2, 231-3, 231-4) said corresponding cells, evolving the probability scores based on the time elapsed since a latest update of these probability scores; and controlling the autonomous vehicle on the basis of the occupancy grid. In some embodiments, the probability score is updated based on a comparison of a measured elevation and an elevation indicated by the reference map data.