Radar Scene Emulator

See the Whole Picture

In vehicle autonomy, advancing through Levels 3, 4 and 5 demands unshakeable confidence in robust ADAS/AD1 algorithms that are ready for road testing and, ultimately, deployment in real-world vehicles. Keysight’s Autonomous Drive Emulation (ADE) platform is the environment emulator for in-lab testing using realistic roadway scenarios, from mundane to one-in-a-million.

Utilizing “total scene generation,” the platform exercises your systems and algorithms by applying time-synchronized inputs to the actual sensors. Its open architecture also closes the loop with your existing hardware-in-the-loop (HIL) systems and 3D modelers. These capabilities create a solution that complements— and fills the gap between—software simulation and on-road testing.

See, Test, and Accelerate

Keysight’s Radar Scene Emulator (RSE) is a core element of the ADE platform. It tests the performance and response of radar-based ADAS/AD software against complex scenarios in a lab environment that includes near and far targets across a contiguous field of view (FOV). The vehicle’s decisions need to be made on comprehensive real-world scenes, not limited by what the test equipment allows them to see

The RSE provides a deterministic environment for in-lab testing of complex scenes that today can be tested only on the road, if at all. With RSE’s industry-first approach, you can accelerate the learning of your ADAS/AD algorithms by testing decisions earlier against complex, repeatable, high-density scenes, with objects either stationary or in motion

To maximize the realism in every scene, the RSE delivers exceptional capability in four areas: field of view, minimum distance, points per object, and interference. For each of these, standard functionality supports the creation of highly realistic scenarios, and available enhancements add greater realism to ADAS/AD testing. It is now possible to emulate real-world driving scenes in the lab with every variation of environmental condition, traffic density, speed, and distance, with all scene objects represented

Close the Technology Gap With Realistic In-Lab Testing

In ADAS/AD development, one of the crucial goals is achieving confidence in the performance and safety of increasingly autonomous systems. Until now, the testing process has suffered from a gap between software simulation and on-road testing. Software simulation is useful and important, but it lacks the ability to properly emulate the responses of the actual sensors and their imperfections. Roadway testing is essential, but it’s costly, nonrepeatable, and too risky to perform too early in the development process. Ultimately, the ability of ADAS/AD algorithms to ensure the safety of people and property is of the utmost importance. For that reason alone, it is critical to utilize all three approaches: software simulation, in-lab scene generation, and on-road testing.

The rendering of full 3D scenes in the lab, with the goal of testing the real sensor response, is the vision. Today, specifically in emulating radar targets, there is a technology gap.

• Limited number of targets and field of view

• Limited ability to test at close range

• Poor object resolution

Backed by our dedicated team of researchers and engineers, the RSE fills this gap with the ability to emulate complex and detailed roadway scenarios in your lab. The key concept is total scene generation, applied consistently and repeatably across a contiguous field of view.

Create Scenarios That Range From Mundane to One-In-A-Million

Validating ADAS/AD radar detection routines against a small number of targets offers an incomplete view of the real world

The RSE allows you to create scenarios containing up to 512 point-targets with independent attributes: speed, direction, distance from the ego vehicle, angle, and more