New: Easy-to-use ROS 2 driver for smartmicro’s automotive sensors
For example, in the video you can see a vehicle driving through the city, equipped with 4 smartmicro UMRR-96 Type 153 short-range radars (detections visualized with blue, orange, purple, and white circles) and 2 smartmicro UMRR-11 Type 132 Automotive long-range radars (detections visualized with green circle). All sensors are sending their high-performance detections through the new ROS 2 driver.
(Video courtesy of Virtual Vehicle)
With the Robot Operating Systems (ROS) community continuing to grow, the demand from our customers to provide a ROS 2 driver which interfaces with our smartmicro sensors has increased. Our ROS 1 driver is CAN-based, which is a widely used protocol in the automotive industry. However, with vehicles typically including multiple sensing technologies like cameras, radars, infotainment systems, and in-dash displays, the demand for bandwidth is tremendous.
To fulfil this need, we are now releasing our new Ethernet-based ROS 2 driver interfacing with smartmicro sensors. We combined the ROS 2 expertise from Apex.AI with the smartmicro know-how on sensor driver development. We tested it with the help of the integration experts at Virtual Vehicle, an R&D center for advanced virtualization of vehicle development.
Monitoring the entire 360-degree surroundings of a vehicle, the performance of our radar sensors is unaffected by dust or other critical environments. This makes them a great fit for any detection system, especially for autonomous driving applications. An important factor for the use of sensors is, of course, the use of their data.
This new ROS 2 driver offers multi-sensor support, connecting up to 10 UMRR-96 Type 153 and UMRR-11 Type 132 Automotive sensors at a time with one ROS 2 driver node. This node is configurable using ROS 2 parameters (e.g., setting the IP address, ID, and interface) and interfaces with our Smart Access C++ API for direct connection to smartmicro sensors. With this, Apex.AI makes the setup process as easy as possible, saving users of smartmicro radars efforts of manual, customer-specific procedures of sensor configuration to connect the sensors via the correct underlying library. Using ROS 2, the sensor data stream can easily be processed and re-published within any system based on ROS 2, without requiring the need of extending the system. Additionally, it saves the user the familiarization effort with our Smart Access API itself. The driver even has integration tests based on official ROS 2 testing methods and gtest to ensure seamless integration into the customer system.
Based on Data Distribution Service (DDS) and implemented in C++, ROS 2 shows a much faster performance and is more resourceful than ROS 1, making it suitable for real-time data distribution to embedded systems. It also enables easier portability between different versions of ROS and Ubuntu and has a lower latency and smaller memory footprint.
The ROS 2 driver enables future innovations in Robot Operating Systems for automotive applications, like Advanced Driver Assistance Systems (ADAS) or Autonomous Driving (AD) using smartmicro sensors. As one example, Autoware.Auto could use smartmicro’s automotive sensors to implement sensor fusion to increase reliability and safety in the perception system.
The driver is hosted on GitHub and made open-source, licensed under MIT, and available for commercial use. This enables any developer or user to get the driver easily and, if needed, make requests and reports. Besides our active support for the driver on GitHub, the open-source availability allows for the expertise of other users from different domains to enhance the ROS 2 driver.
Read more or download the ROS 2 driver.