Ibeo Automotive Systems, an automotive LiDAR specialist based in Hamburg, has developed a new automation solution to validate ADAS and the sensors required for autonomous driving.
The ibeo.Reference toolchain also automates and specifies the process of manual labeling of data. It has always been a very human and time consuming task. You can refer to almost all traditional sensor systems.
ADAS and AD must be validated in a complex process before they are finally fitted to the vehicle and ready for continued production. Regardless of the technology of the sensor system (camera, radar or LiDAR), baseline data is required for each system and should be compared to data acquired during the testing process.
The Ibeo recording system is installed in the vehicle along with the manufacturer’s prototype system to record data in parallel. It can also be used to create scenarios for the development of ADAS or AD algorithms.
The recorded data collected on the verification drive can last for thousands to average hours. Additionally, a manual assessment will generate up to 300 hours of tagging activity for every hour of on-road testing.
“Manually adjusting this data is currently a time consuming and costly factor in many new developments,” said Philip Neitzel, product manager at ibeo.Reference.
The industry has traditionally used outsourced service providers to categorize the large amounts of recorded data. Often, hundreds of employees will manually evaluate each meter of road to see if the sensor under test detects a car, person, lane markings or guardrail, for example to adjust the data. Judge.
Ibeo’s new toolchain, the ibeo Validation Bundle, is working on this task. “As part of the ibeo validation pack, we provide a comprehensive automated solution in the form of ibeo automatic annotation software that algorithmically generates reference data from the incoming data,” added Neitzel. ..
“If you want, we can process it manually to generate ground truth data, data that is a realistic and complete representation of the vehicle environment. In this way, one of the major challenges we face today. To the series products. “
With the new ibeo.Reference toolchain, users receive the building blocks needed to validate sensors from a single source.
- A recording system composed of hardware and software running on reference data. This is the first time that some third-party LiDAR sensors are also supported.
- As a scalable, cloud-compatible software for automatically creating labeled reference data, ibeo Auto Annotation is an editor for further processing of reference data into ground truth data.
“It was very important for us to cover almost all of the customer challenges in how to combine LiDAR components and deliver a complete system that is technically scaled up on a regular basis,” said Neitzel. Declared.
“In this way, we help our customers grow faster, more accurately and more profitably, and we are much closer to the overall vision of fully automated mobility.
The ibeo.Reference toolchain is now available. In addition, Ibeo offers a test license, which you can request from support @ ibeo-as.com.
See the press release
Original source: https: //www.ibeo-as.com/en
Source link Ibeo develops the verification of ADAS sensors | Automotive Industry News