Expected Contour Computation for Collision Avoidance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for collision avoidance in automated devices, such as vehicles, are inefficient due to the need for extensive computations to account for system, localization, and perception uncertainties, particularly when ensuring a high confidence of no collision, which is time-consuming and demanding on control unit performance.
Innovation Solution
A method that computes a rapid 'belief print' by determining a base polyhedron with a limited number of corners, transforming it based on probability densities, and forming an expected contour to ensure collision avoidance with static and dynamic surroundings, incorporating uncertainties and confidence intervals, allowing for efficient collision checks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If extensive collision checks are performed to ensure high confidence of no collision, then reliability of collision avoidance is improved, but computation time and processing demands increase significantly
Solution Approach 1:
The patent segments the continuous probability distribution into a discrete set of sigma points (typically 7 points for 3D state). Instead of performing collision checks across the entire continuous distribution, the system checks only at these discrete sigma points, dramatically reducing computation while maintaining statistical accuracy through weighted combination of results.
Solution Approach 2:
The patent transforms the collision check problem from verifying continuous probability distributions to evaluating discrete sigma points with associated weights. By changing the parameter representation from continuous coordinates to discrete sigma points with weights, the system achieves both speed and accuracy in collision assessment.
2Productivity
If sampling methods are used to check collision probability, then computational demands are reduced, but the ability to ensure strict confidence intervals is lost
Solution Approach 1:
The patent introduces sigma points as intermediary representations that bridge the gap between continuous probability distributions and discrete sampling. These sigma points are specifically designed to capture the essential statistical moments (mean, covariance) of the distribution while enabling efficient discrete evaluation, thus maintaining both speed and statistical rigor.
3Measurement precision
If the footprint is transformed to multiple sigma points for collision checking, then collision detection accuracy is improved, but the number of required checks increases exponentially
Solution Approach 1:
The patent applies partial action by selecting only the essential sigma points (7 points for 3D state) that sufficiently represent the probability distribution, rather than checking all possible configurations. This partial sampling approach maintains adequate collision detection accuracy while avoiding exponential complexity growth.
Data Source
AI summary
A method for ascertaining an expected contour of a mobile or stationary device for avoiding collisions, using at least one control unit that is internal external to the device includes: obtaining a movement trajectory of the device, which contains probability densities based on state estimation, at least based on expected values and covariances; obtaining a base polyhedron and an approximate contour of the device having a limited number of corners, a confidence interval within which a collision with the static and dynamic surroundings of the device is to be avoided being defined; transforming the base polyhedron to the at least one probability density of the movement trajectory that describes the state estimation; and, for each corner of the transformed base polyhedron, computing a transformed device contour, and ascertaining the expected contour of the device with inclusion of all transformed device contours.