TESTING Frost bytes Measuring perception sensor performance in winter conditions By Mike Dempsey , Claytex View of the winter test site with sensors housed in the black box in the foreground and targets distributed across the site BOOTH: 6200 A dverse weather conditions such as rain, snow, fog and hail can greatly impair the performance of AV sensors. For example, raindrops and snowflakes can distort the signals of lidar sensors, and fog can reduce visibility for cameras, leading to incorrect object recognition. This sensor degradation in poor weather poses a significant challenge to the safety and effectiveness of autonomous vehicles. The Sim4CAMSens project, part-funded by Innovate UK and led by Claytex, aims to revolutionize the field of autonomous vehicle technology by developing a robust supply chain for the modeling, simulation, testing and characterization of the perception sensors used in automated driving functions. At the heart of Sim4CAMSens is a collaboration between a consortium of expert partners, each contributing unique expertise to address the challenges being faced. Sim4CAMSens targets three key sensor technologies: radar, camera and lidar. These sensors are essential for the safe and efficient operation of AVs, as they provide vital data about the vehicle’s surroundings. The performance of these sensors is therefore a critical safety issue. To investigate how it varies in winter conditions, Claytex designed and executed a winter test campaign focused on capturing data that will enable the project partners to understand the relationship between weather and sensor performance. As with any real-world test work, one of the key concerns was minimizing the number of variables that could affect the measurements. The fewer the number of variables, the easier it is to find relationships between the performance metrics and remaining variables. With the weather conditions being the primary variables of interest for this test campaign, it was decided to make the sensors and targets static. The sensor suite included four cameras with three different CMOS chips, five lidar covering 850nm, 905nm and 1,550nm wavelengths, five radars all operating within the automotive 77-79GHz band and a thermal imaging camera. Some of these are commercially available sensors while others are development kits that provide access to raw sensor data and, hopefully, greater insights into the physics affecting the performance of each sensor. A bespoke sensor housing has been designed to protect the sensors and dataloggers throughout the winter. It features four windows on the front made from polycarbonate that is Front view of the sensor enclosure transparent to the cameras, radar and lidar housed inside. The rest of the housing is constructed from steel, fully insulated, and includes temperature and humidity controls to ensure the interior stays warm and dry. Underneath the housing there is an array of four lidar sensors and the thermal imaging camera. By placing these sensors underneath the main housing, they are shielded from the rain and snow. This means that any change in performance is related to the effect of weather in the atmosphere rather than droplets on the surface of the sensor. Measuring and quantifying the weather is one of the biggest challenges because of the huge uncertainty associated with any weather measurement and the large time and spatial variations that can occur even over a relatively short period and small test site. The suite of weather instruments covers the basics such as temperature, pressure, humidity and rain rate and is supported by specialist instruments including a disdrometer and present weather sensor. Lidar view of Analysis of the data collected during the test site the test campaign is now underway, leveraging the expertise of the project partners to start quantifying the performance of each sensor under different weather conditions. This information will be used to help improve the simulation tools, measurement processes and perception sensors developed by the partners. To learn more about this project and the next steps for Sim4CAMSens, please visit avsandbox.com and get in touch with the team at Claytex. To find out more, scan the QR code or visit: www.www.claytex.com Claytex 72 ADAS & Autonomous Vehicle International April 2024