Automotive Radar Testing and Big Data: Safeguarding the Future of Driving

In today's automotive landscape, Advanced Driver Assistance Systems (ADAS) is becoming ubiquitous in modern vehicles. It's become a valuable tool for drivers, helping them stay safe on the road. The key enabler of ADAS is automotive radar, the technology that scans the road, detects obstacles, and makes split-second decisions to maintain safety.

However, radar sensors must be tested meticulously to ensure they perform flawlessly. That's where automotive radar end-of-line (EOL) testing steps in.

Testing the Radar

EOL testing is paramount in the radar manufacturing process. It is the final checkpoint before radar modules are installed. EOL testing ensures radar sensors are calibrated correctly, aligned accurately, and functioning as they should.

Field of view (FOV) tests are performed to guarantee that radar sensors can "see" objects on the road accurately. FOV tests are conducted using an automated radar test system that integrates a radar target simulator (RTS) with an anechoic chamber to enable testing of automotive radars over the air. This type of test system can simulate objects of various sizes correlating to real objects such as pedestrians, cars, or trucks. In addition, the test system can simulate objects at various angles and distances to test the radar performance in a factory or lab environment.

During FOV tests, large amounts of data is collected from the radar and the test system. The data is often archived, but with the help of big data analytics, it can be turned into actionable insights to create additional value.

How Big Data Enables Manufacturing Excellence

Big data analytics helps radar manufacturers analyze measurement trends and view critical metrics, ensuring quality control and process optimization. Using analytics software, engineers can fine-tune their processes and ensure that radar sensors perform at their best.  Here's how big data analytics can empower radar test engineers:

· Parsing and processing data to generate meaningful, high-level views of critical metrics.

· Tracking and achieving production targets through real-time views of critical metrics, including overall equipment effectiveness, yield, first-pass yield, retest, and volume.

 · Gaining insights and driving timely actions by predicting issues based on historical measurement trends. For example, monitoring the process capability index (CPK) to alert users about any trend indicating a drop in quality.

 · Reducing downtime by identifying anomalies and providing insights into root causes by analyzing correlations between variables.

 · Managing capacity based on production line utilization, yield, and equipment availability.

 How Analytics Reshapes EOL Testing

In radar EOL testing, one of the key verification parameters is the radar cross-section (RCS) detection accuracy, which represents the size of an object. Unlike passive objects that have fixed RCS, RTS allows the simulation of various levels of RCS, echoing a desired object size for radar detection. 

While RTS systems offer versatility for radar testing, they present challenges to overcome. One such challenge is the sensitivity of the system’s millimeter-wave (mmWave) components to temperature variations, which can significantly impact the ability to accurately simulate RCS values. Therefore, controlling the ambient temperature in a testing setup is important to ensuring that the RTS replicates the RCS expected for a given object size.

Let's delve into a scenario that underscores the importance of analytics software in maintaining measurement quality and addressing anomalies in a radar EOL test.

For example, a trend analysis of RCS measurements over time reveals a concerning deterioration. Several anomalies are flagged in the data, indicated in pink in Figure 2. This degradation in the measurement trend also corresponds to a drop in the CPK value over a six-hour period. Depending on the predefined limits and criteria, the analytics software is programmed to trigger real-time alerts to the end user. These alerts include diagnostic details such as information about the equipment, fixtures, and the specific test being conducted. This proactive alerting mechanism empowers the end-user to promptly intervene and rectify any issues, thereby mitigating potential risks to product quality.

 Once the user is alerted to the presence of an anomaly within the testing system or process, identifying the root cause of this anomaly becomes the next critical step. This is where manufacturing analytics software plays a pivotal role by identifying root causes.

Figure 3 reveals a correlation uncovered through process monitoring and the analytics software, illustrating a connection between the increase in RCS distribution and variations in RTS temperature. This insight provides a crucial clue to understanding the anomalies in the data. Figure 4 offers a closer look at the specific data points identified as anomalies, accompanied by the notification message generated by the analytic software.

Through this analysis, it was determined that a malfunctioning ventilation port within the testing facilities was responsible for the heat buildup around the system. This led to a significant temperature increase of more than 10°C within a single day of testing.

In a production line scenario without the early detection capabilities provided by analytics software, the delay in identifying a deteriorating RCS distribution can be profound. First, the delay can result in the production of a number of radar sensor modules with compromised quality. With each unit requiring a minute of testing, this can mean a significant amount of retest time is needed to assess the sensor modules. Not only does this reduce the overall output capacity of the production line, but it also translates into considerable labor and resource costs.

Furthermore, the repercussions extend beyond the immediate operational setbacks with. the need to scrap a number of radar faulty module units. Not only does this represent a direct monetary loss and the overall profit margin, but it also contributes to waste and environmental concerns.

All these adverse outcomes, from reduced output capacity to financial losses and environmental impact, highlight the critical importance of integrating analytics software into an automotive radar EOL testing solution. This software serves as an invaluable tool for identifying and rectifying potential issues far earlier than human intelligence alone can achieve. By doing so, it enables manufacturers to proactively address quality issues, minimize production delays and safeguard their bottom line. In the fast-paced and demanding landscape of modern production, such technological advancements are not just advantageous but often essential for maintaining efficiency, profitability, and product integrity.

Conclusion

In conclusion, analytics software, particularly in the context of radar EOL testing, is proven to be indispensable for maintaining measurement quality, detecting anomalies, and expeditiously identifying root causes. By leveraging data analytics and correlations, anomalies in the production process can be quickly identified, and the solution to rectify the anomalies can be quickly formulated. In addition, it empowers users to make informed decisions to safeguard product quality and company revenue in critical industries like automotive radar manufacturing.

-ByKeng Fai, an R&D engineer at Keysight Technologies & Pereddy Vijai Krishna Reddy, Data scientist, Keysight Technologies