Vehicle Object Recognition and Motion Prediction Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing vehicle systems face increased calculation loads when predicting the future position of multiple objects around a vehicle, which can lead to inefficiencies in prediction accuracy and processing load, especially when global optimization is performed without considering individual program functions.
Innovation Solution
A vehicle system that performs individual optimization of recognition and motion prediction programs separately, followed by global optimization, to reduce processing load and enhance prediction accuracy by leveraging specialized optimizations for each program before executing global optimization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If global optimization is performed from the beginning for multiple programs, then system-wide optimization is achieved, but processing load increases significantly
Solution Approach 1:
The optimization process is segmented into two distinct phases: individual optimization for each program (recognition program and motion prediction program) followed by global optimization. This segmentation reduces the processing load by breaking down the complex global optimization problem into smaller, more manageable individual optimization tasks, while still achieving system-wide optimization benefits in the second phase.
2Productivity
If individual optimization is performed for each program, then processing load is reduced, but system-wide optimization is not achieved
Solution Approach 1:
Individual optimization is performed as a preliminary action before global optimization. This preliminary optimization of each program reduces the processing load and prepares the system for the subsequent global optimization phase, ensuring that both local and system-wide optimization goals are achieved without excessive computational burden.
3Adaptability or versatility
If the number of objects or object types is increased, then recognition capability is improved, but calculation load increases
Solution Approach 1:
The recognition and prediction tasks are segmented into separate programs with specialized functions. The recognition program handles object identification while the motion prediction program handles future position prediction. This segmentation allows the system to maintain high recognition capability for multiple object types while managing calculation load through specialized, optimized processing for each program.
Data Source
AI summary
A vehicle system includes an electronic control unit. The electronic control unit is configured to execute a first program, a second program, and a third program. The first program is configured to recognize an object present around a vehicle, the second program is configured to store information related to the recognized object as time-series map data, and the third program is configured to predict a future position of the object based on the stored time-series map data. The first program and the third program are configured to be (i) first, individually optimized based on first training data corresponding to output of the first program and second training data corresponding to output of the third program, and (ii) then, collectively optimized based on the second training data corresponding to the output of the third program.