LIDAR Obstacle Model Using Shadow Regions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current obstacle data models using light-based range sensors fail to accurately capture the rigidity of underlying surfaces, leading to inefficient motion planning and potential penetration into solid obstacles, as they do not effectively utilize sensor shadows to estimate obstacle extent.
Innovation Solution
The method involves scanning the obstacle space with a LIDAR sensor to generate boundary data sets, deriving shrouded regions, and identifying high-confidence occupancy regions by intersecting these regions, with a decay factor adjusting the data sets to account for movement and confidence levels, allowing for adjustments in mission plans to avoid obstacles and limit further scanning in high-confidence areas.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If light-based range sensors are used to construct obstacle data models, then the system can detect surface information, but it fails to capture the rigidity of underlying surfaces leading to inaccurate obstacle extent estimation
Solution Approach 1:
The patent converts the harmful effect of sensor shadows (which normally represent missing or uncertain data) into a beneficial source of information about obstacle extent. By analyzing the boundaries of shrouded regions where sensor beams are blocked, the system infers the spatial boundaries of obstacles, transforming an information gap into a meaningful measurement of obstacle geometry and rigidity.
Solution Approach 2:
The patent introduces shrouded region boundaries as an intermediary representation between raw sensor hits and obstacle rigidity estimation. These boundaries serve as a mediator that encodes indirect information about obstacle extent, allowing the system to infer rigidity properties without directly measuring them, thus bridging the information gap created by light-based sensing limitations.
2Reliability
If the system uses only surface information for motion planning, then planning computation is simpler, but the vehicle may compute paths that penetrate solid obstacles
Solution Approach 1:
The patent performs preliminary action by pre-computing shrouded regions and their boundaries from sensor data before motion planning occurs. This advance processing creates a enriched obstacle representation that includes extent and rigidity information, allowing the motion planner to query pre-processed spatial data rather than computing obstacle properties during real-time path planning, thus improving reliability without proportionally increasing real-time computational complexity.
3Reliability
If subsequent exposures are used to resolve obstacle solidity, then obstacle rigidity information may be obtained, but planning efficiency decreases and avoidance reactivity is delayed
Solution Approach 1:
The patent applies preliminary action by computing shrouded regions and obstacle extent information in advance from available sensor data, rather than waiting for subsequent exposures. This pre-computation enables the system to make informed planning decisions immediately, improving both planning efficiency and avoidance reactivity while maintaining reliable obstacle solidity detection.
Solution Approach 2:
The patent provides beforehand cushioning by creating a buffer of pre-computed obstacle extent and rigidity information from shrouded region analysis. This preparatory information cushion allows the motion planning system to make reliable decisions without requiring additional time for subsequent sensor exposures, thus compensating for potential information gaps in advance and maintaining both reliability and temporal efficiency.
4Measurement precision
If the system scans all areas continuously to ensure complete obstacle detection, then detection coverage is maximized, but resource consumption increases
Solution Approach 1:
The patent extracts and utilizes valuable obstacle extent information from shrouded region boundaries, which are byproducts of normal sensor operation. By deriving meaningful data from these previously unused shadow regions, the system improves detection coverage without requiring additional scanning resources, as the shrouded region analysis leverages existing sensor beams and their blockage patterns.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enhances motion planning by accurately estimating obstacle extent using sensor shadows, reducing the likelihood of penetrating solid obstacles and optimizing resource usage by adjusting mission plans and reducing unnecessary scanning.
Implementation Method 1
Light-based range sensors operate by returning the range of a reflecting surface they 'shine' on
Implementation Method 2
scanning the vehicle obstacle space at the first and second positions, generating first and second boundary data sets from results of the scanning... the scanning includes conical scanning ahead of the vehicle, a cone-shape of the conical scanning being defined by a field of view of a light detection and ranging (LIDAR) sensor
Data Source
AI summary
A method of operating an obstacle data model construction system of an aircraft is provided. With a vehicle moving through a vehicle obstacle space from a first to a second position, the method includes scanning the vehicle obstacle space at the first and second positions, generating first and second boundary data sets from results of the scanning at the first and second positions, respectively, and deriving first and second shrouded regions from the first and second boundary data sets, respectively. The method further includes identifying a high confidence occupancy region from intersecting portions of the first and second shrouded regions or identifying an occupancy region from a union of the first and second shrouded regions.