Denso packs radar and vision in global safety product
The Global Safety Package has been developed in cooperation with J-QuAD and uses the combined performance of a millimeter-wave radar sensor and vision sensor to assist the driver in controlling the vehicle safely. This radar sensor detects the shapes of road objects, such as vehicles and guardrails, while the vision sensor uses a camera to detect the environment ahead of the vehicle.
With a wider detection angle, longer detection range, and improved speed resolution, the millimeter-wave radar sensor helps avoid collisions at intersections, whether with other vehicles or pedestrians. Thanks to its improved detection capability and reduced size and cost, this radar sensor has fewer installation requirements and can be used in more vehicle types and models, the vendor promises.
The sensor has been improved over the predecessor under various aspects: It features a wider detection angle and a longer detection range with improvements such as in-antenna design and radio wave transmission efficiency. Through the increased frequency modulation of radio wave transmissions, the speed resolution has been improved. Last but not least, the new radar sensor also is smaller and more cost-effective, the it is using fewer than half the number of parts compared to the previous generation. Denso achieved this by integrating the antenna and power supply board, simplifying product structure and leveraging multiple-input and multiple-output (MIMO) technology, which allows for a smaller antenna with fewer channels to transmit and receive radio waves.
The vision sensor has a wider horizontal field of view to help with collision avoidance at intersections, as well as a longer range to detect road objects in front of the vehicle. This assists adaptive cruise control and lane changes over a broader speed range. Moreover, the product uses artificial intelligence (AI) to improve sensing performance and assistance functions, such as headway distance control – the ability to maintain consistent distance between a preceding vehicle – and traffic sign recognition.
Points of product development include a high-sensitivity, high-resolution imager with a wider horizontal field of view and longer detection range and a reduced product size and cost by integrating image processing and system control functions, which reduced the number of microcomputers used; adjusting the structural design to control heat emitted from the processor; and reviewed the lens design.
What’s more, AI technology is used to identify drivable areas, preceding vehicles, road signs, and other solid objects based on image data. Vehicles and pedestrians in all directions, including those ahead, behind, and to the side, can be detected.