Travel Command Generation Device Using Collision Probability Table
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current collision avoidance systems in advanced driving assistance technologies face increased computation time and potential control cycle delays when calculating collision probabilities for multiple speed profiles, especially in automatic travel control scenarios.
Innovation Solution
A travel command generation device that computes collision avoidance commands by generating a collision probability table based on obstacle existence probability distributions, using input movement distance and time, and searches for travel commands that minimize collision probabilities, allowing for parallel computation and efficient storage in compact memory.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If iterative computation is performed to search for the optimum speed profile that minimizes collision probability, then collision avoidance accuracy is improved, but computation time increases and control cycle completion becomes uncertain
Solution Approach 1:
The collision probability table is pre-computed for various movement distances and times before the actual travel command generation. This preliminary computation stores collision probabilities for different scenarios in advance, allowing the system to quickly retrieve and compare pre-calculated values during iterative optimization without performing full computations each cycle, thus reducing real-time computation time while maintaining accuracy
Solution Approach 2:
The invention creates a simplified representation of collision probability data in the form of a lookup table that copies essential computation results. Instead of recalculating complex collision probabilities during each iterative search, the system uses pre-computed table entries that replicate the essential information needed for optimization, significantly reducing computational burden while preserving the accuracy needed for finding optimal speed profiles
2Reliability
If multiple speed profiles are evaluated to compute collision probability in automatic travel control, then collision avoidance reliability is improved, but computation complexity increases
Solution Approach 1:
The collision probability table is pre-computed for various movement distances and times before the actual travel command generation. This preliminary computation stores collision probabilities for different scenarios in advance, allowing the system to quickly retrieve and compare pre-calculated values during iterative optimization without performing full computations each cycle, thus reducing real-time computation time while maintaining accuracy
Solution Approach 2:
The invention creates a simplified representation of collision probability data in the form of a lookup table that copies essential computation results. Instead of recalculating complex collision probabilities during each iterative search, the system uses pre-computed table entries that replicate the essential information needed for optimization, significantly reducing computational burden while preserving the accuracy needed for finding optimal speed profiles
Data Source
AI summary
Provided is a travel command generation device that generates, on the basis of an existence probability distribution for a plurality of obstacles, travel commands to avoid collision between the plurality of obstacles and a host vehicle, wherein the speed of collision probability computation is enhanced. On the basis of the existence probability distribution for the plurality of obstacles, a collision probability table is generated, for which a movement distance L on a fixed trajectory and a time T are input and the probabilities of collision between the plurality of obstacles and the host vehicle are output; and on the basis of the collision probability table, travel commands are generated for avoiding collision between the plurality of obstacles and the host vehicle.


