Sensor Test Data With Introductory Scenarios for Plausible ADAS Startup
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional simulation systems for ADAS/AD systems fail due to implausible initial scenarios, leading to test failures when synthetic data is used, as objects in the surroundings appear, disappear, or move unnaturally, causing the control unit to abort the test.
Innovation Solution
Generate test data by supplementing real-world recorded data with a quasi-continuous introductory scenario using synthetic objects to create a plausible initial state, ensuring seamless transition into the original scenario, thereby avoiding discontinuities and implausibilities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If synthetic sensor raw data is used to stimulate control units, then test coverage and safety validation can be achieved, but the initial scenario may be implausible causing objects to appear, disappear, or move unnaturally which leads to test failures
Solution Approach 1:
The patent applies preliminary action by generating an introductory scenario before the main test scenario. This introductory scenario includes synthetic sensor data that gradually introduces objects into the scene, allowing the control unit to adapt to the test environment before actual testing begins. The introductory scenario prepares the system by establishing a plausible initial state with objects appearing naturally, thereby avoiding test failures caused by abrupt or implausible object appearances in the main scenario.
2Measurement precision
If recorded real-world data is used for testing, then realism and accuracy are improved, but the data must be synchronized and adapted in real-time which increases processing complexity
Solution Approach 1:
The patent applies segmentation by dividing the test data into two distinct segments: recorded real-world sensor data and synthetic introductory scenario data. Each segment is processed separately with appropriate synchronization and adaptation techniques. The real-world data maintains its original high accuracy while the synthetic data provides a controlled introductory context. This segmentation allows the system to leverage the advantages of both data types without the full complexity of processing them uniformly.
3Productivity
If the sensor starts moving at full speed in the recorded data, then the recording period can be shortened, but the initial state becomes implausible and causes control units to abort the test
Solution Approach 1:
The patent introduces an intermediary solution by inserting synthetic introductory scenario data between the start of the test and the actual recorded real-world data. This intermediary data acts as a bridge that gradually transitions the sensor from a stationary state to the moving state recorded in the real-world data. The control unit experiences a plausible acceleration sequence through the synthetic data before encountering the full-speed recorded data, thereby accepting the test while maintaining recording efficiency.
Data Source
AI summary
A method for processing raw data recorded from the real world with the aid of a sensor into test data for stimulating a control unit, including: providing recorded raw data from the real world; ascertaining real objects detected by the sensor from the raw data, and generating route data sets, which each describe a scene including images of these real objects at consecutive points in time. providing a library of synthetic objects; assigning synthetic objects to detected images of the real objects, replacing the detected images with synthetic objects; supplementing temporally consecutive route data sets with supplementary data sets before the first route data set; and generating test data by converting the route data sets supplemented by the supplementary data sets into raw data that would have been recorded in the real world by the sensor during the introductory period and the recording period.