Learned Model Grid Map Correction for Blind Area Object Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional techniques face difficulties in accurately estimating objects at non-observation positions, especially when the result of tracking other vehicles is not used or the object has never been observed before, leading to uncertainty in determining the presence or absence of objects in blind areas.
Innovation Solution
An information processing device that acquires and correlates observation information across grids in a particular space using a learned model, correcting correlations to improve estimation accuracy by integrating object and non-observation information, enabling more precise prediction of object presence even in unobserved areas.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional tracking techniques are used to estimate objects in blind areas, then estimation can be performed using available sensor data, but accuracy deteriorates when tracking results are unavailable or objects have never been observed before
Solution Approach 1:
The patent introduces a learned model as an intermediary between sensor observation data and object estimation results. This model processes observation information from multiple grids and temporal sequences to generate accurate estimates even when direct tracking data is unavailable, resolving the contradiction between reliability and precision in unobserved scenarios
Solution Approach 2:
The system performs preliminary actions by continuously acquiring and storing observation information in advance across multiple grids and time steps. This pre-collected data serves as the foundation for accurate estimation when objects enter blind areas, enabling reliable prediction without requiring real-time tracking of every object
2Measurement precision
If sensor coverage is expanded to observe all areas, then object detection accuracy improves, but device complexity and cost increase significantly
Solution Approach 1:
The patent transitions from spatial expansion (adding more sensors) to temporal and contextual dimension expansion (using learned models to process observation information across time and multiple grids). This allows accurate object estimation in blind areas without physically expanding sensor coverage, maintaining simplicity while improving detection accuracy through intelligent processing of existing data
Data Source
AI summary
According to an embodiment, an information processing device includes one or more processors. The one or more processors is configured to acquire a map in which, for each of grids in a particular space, observation information representing object information on an object or the observation information representing non-observation information on non-observation of the object is correlated; and correct, for each of the grids, correlation of the observation information by using a learned model based on the observation information correlated with other peripheral grids.