Sensor Visibility Estimation Using Bounding Boxes and Ray Sampling

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing sensor algorithms for determining object visibility are computationally intensive, making them unsuitable for real-time applications requiring high accuracy.

Innovation Solution

A method using oriented bounding boxes and a simplified ray tracing approach to estimate object visibility, employing a predefined sensor position, constant angle differences between rays, and a two-counter system to determine visibility percentages with reduced computational effort.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If high accuracy algorithms are used to determine object visibility, then measurement precision is improved, but computational time increases

Engineering Contradiction:
Improvevisibility determination accuracyVSAvoidcomputational time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent segments the continuous field of view into discrete rays with constant angular intervals. By dividing the spatial domain into manageable ray segments, the algorithm can efficiently sample visibility without requiring exhaustive computation across the entire continuous space, thus maintaining accuracy while reducing computational time.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent uses a two-counter system where the first counter tracks total ray intersections and the second counter tracks visible surface intersections. By using partial sampling through rays rather than complete surface analysis, the algorithm achieves sufficient visibility estimation without performing excessive computational actions, balancing accuracy and speed.

Inventive Principle:
Principle #16Partial or excessive action

2Measurement precision

If complex ray tracing algorithms are used to achieve accurate visibility estimation, then measurement precision is improved, but device complexity increases

Engineering Contradiction:
Improvevisibility estimation accuracyVSAvoidalgorithm complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent changes the approach by using constant angular intervals for rays and implementing a two-counter system instead of complex continuous rendering algorithms. This parameter change simplifies the algorithmic structure while maintaining visibility estimation accuracy, reducing device complexity without sacrificing measurement precision.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent uses simplified geometric representations (oriented bounding boxes) that copy the essential spatial properties of objects without requiring complete geometric models. This copying approach maintains visibility estimation accuracy while significantly reducing the complexity of geometric processing required.

Inventive Principle:
Principle #26Copying

Data Source

PatentEP3722976B1Method for estimating visibility of objects
Publication Date: 2026.04.01 APTIV TECHNOLOGIES LTD
  • EP3722976B1 patent drawingFigure 1
  • EP3722976B1 patent drawingFigure 2
  • EP3722976B1 patent drawingFigure 3

AI summary

A computer implemented method is provided for estimating whether an object is visible for a sensor. A position and an orientation of at least two objects are determined and surfaces are defined which are tangent to the respective object and which cover the respective object. A plurality of rays is defined starting at a predefined sensor position which specifies an origin of a three-dimensional coordinate system. Each ray has an azimuth angle and an elevation angle with respect to the coordinate system. A difference of respective azimuth and elevation angles of each pair of adjacent rays is constant. In addition, the plurality of rays covers a region in which the at least two objects are located. For each object, a respective subset of rays is identified which have a point of intersection with at least one of the surfaces of the respective object, and a number of rays of the subset is determined. A respective visibility index is associated to each object, and each visibility index is initialized with zero. For each ray of the subsets, the object comprising the point of intersection having the shortest distance to the origin of the coordinate system is identified, and the visibility index for the identified object is increased by one. A percentage of visibility for each object is determined as a ratio of the visibility index associated to the respective object and the number of rays of the subset of the same object.