Abstract: This document describes techniques and systems related to a radar system using a machine-learned model to identify the number of objects within each range-Doppler bin. For example, the radar system includes a processor that obtains radar data associated with objects and processes the radar data to generate beam vectors. The processor then uses a machine-learned model to identify the number of objects within each range-Doppler bin using extracted features (e.g., magnitude variation, signal-to-noise ratio, subarray beam vector correlations) of the beam vectors. The processor selects a particular angle-finding technique based on whether a single object or multiple objects are identified. In this way, the described systems and techniques more accurately identify the number of object in each range-Doppler bin, thus improving the computational efficiency and robustness of subsequent angle finding.

Abstract: A computer-implemented method includes receiving environment description data associated with a simulated environment including an object and identifying an object type of the object. The method includes, in response to the object type being a human, identifying a motion type associated with the object, loading a CAD model associated with the object, loading a motion file associated with the motion type, and mapping the CAD model to the motion file. The method includes performing ray tracing simulation, performing physical optics simulation, calculating at least one of a Doppler shift and a micro-Doppler shift for each ray of a set of rays, performing ray clustering, and transforming a simulation output for display on a user device. The simulation output includes a motion simulation associated with the motion type of the object. The motion simulation includes a main motion of the object and a set of micro-motions of the object.

Abstract: This document describes techniques and systems for accurate and efficient electromagnetic response for a sensor simulator. Target information and sensor parameters for an electromagnetic sensor are simulated in an environment that includes a ground plane. Electromagnetic rays that may be detected by the sensor or an image of the sensor are launched from the simulated sensor toward the target and an image of the target about the ground plane to determine a complex electromagnetic response of the target. A ray-tracing algorithm is applied to trace the forward wave propagation of electromagnetic rays in the environment that considers rays bouncing between the target and the image of the target. An electromagnetic response can be modeled based on the congregation of the electromagnetic response of all backward paths of all bounces of all rays. In this manner, an efficient and accurate electromagnetic response model may be approximated.

Abstract: This document describes techniques and systems for electromagnetic response simulation for arbitrary road surface profiles. An electromagnetic response simulator receives a position and an orientation of both an electromagnetic sensor (e.g., radar sensor) and a target, and a geometric profile of a road surface. The road surface may vary in elevation in the lateral and/or longitudinal directions. The electromagnetic response simulator estimates reflection points of electromagnetic rays along the geometric profile of the road surface and translates the positions and the orientations of the electromagnetic sensor and the target into respective local coordinates corresponding to each reflection point. The electromagnetic responses can then be calculated, corresponding simulated rays can be output to a sensor simulator.

Abstract: This document describes techniques and systems related to a radar system using a machine-learned model to identify the number of objects within each range-Doppler bin. For example, the radar system includes a processor that obtains radar data associated with objects and processes the radar data to generate beam vectors. The processor then uses a machine-learned model to identify the number of objects within each range-Doppler bin using extracted features (e.g., magnitude variation, signal-to-noise ratio, subarray beam vector correlations) of the beam vectors. The processor selects a particular angle-finding technique based on whether a single object or multiple objects are identified. In this way, the described systems and techniques more accurately identify the number of object in each range-Doppler bin, thus improving the computational efficiency and robustness of subsequent angle finding.

Abstract: A real-time automotive radar simulation tool is developed based on reduced statistical models summarized from physical-based asymptotic and full-wave simulations. Some models have been verified with measurements. The simulation tool can help save cost and time for the automotive industry, especially for autonomous vehicles. The simulation tool can also help develop new functionalities like target identification or classification as well as help prevent false alarms.

Abstract: This document describes techniques and systems for super-resolution based on iterative multiple-source angle-of-arrival estimation. Beam vectors received by an electromagnetic sensor may include information about multiple objects, but if the objects are close, the objects may initially appear as a single object in a Doppler-range bin. Performing iterative operations on a first angle derived from the beam vector and a subsequent second angle, associated with a second object, derived from the first angle, the first angle and the second angle may be refined and converge toward their actual respective values. The iterative operations include performing calculations involving only the first angle value and the second angle value as unknowns. Noise has been approximated to be random Gaussian noise with zero mean. Additionally, phase ambiguity, associated with sparse channel arrays has been eliminated.

Abstract: This document describes techniques and systems for an adaptive ray-launcher for an electromagnetic response simulator. An adaptive ray-launching process is used to shoot electromagnetic rays at targets in a simulated environment. Uniformly distributed sparse electromagnetic rays are launched at the targets and the angular ray densities relative to the targets are calculated. Several propagation paths that include multipath effects are considered to determine the angular ray densities. Based on the length of the propagation paths of sparse electromagnetic rays with multipath related to each target, denser electromagnetic rays can be launched. The denser electromagnetic rays enable the fidelity related to targets in a far-range to be similar to the fidelity of closer targets. Additionally, any sparse electromagnetics that cannot be detected by a sensor can be disregarded. In this manner, an efficient and accurate electromagnetic response model may be approximated.

Abstract: This document describes techniques and systems for accurate and efficient electromagnetic response for a sensor simulator. Target information and sensor parameters for an electromagnetic sensor are simulated in an environment that includes a ground plane. Electromagnetic rays that may be detected by the sensor or an image of the sensor are launched from the simulated sensor toward the target and an image of the target about the ground plane to determine a complex electromagnetic response of the target. A ray-tracing algorithm is applied to trace the forward wave propagation of electromagnetic rays in the environment that considers rays bouncing between the target and the image of the target. An electromagnetic response can be modeled based on the congregation of the electromagnetic response of all backward paths of all bounces of all rays. In this manner, an efficient and accurate electromagnetic response model may be approximated.

Abstract: This document describes techniques and systems for a modified ray-tracer for an electromagnetic response simulator. Electromagnetic ray information, including a starting point and direction, is received. A potential target can be determined to be hit by the electromagnetic ray by converting the electromagnetic ray information from a global coordinate system of the environment to a local coordinate system of the potential target. The potential target is hit by the electromagnetic ray if a facet of the potential target is computed to be hit by the ray. The computations, performed in the local coordinate system of the potential target, include a simplified large element physical optics formulation for parallel rays. An electromagnetic response related to the potential target can be calculated if the facet of the potential target was determined to be hit. In this manner, an efficient and accurate electromagnetic response model may be approximated.

Abstract: A real-time automotive radar simulation tool is developed based on reduced statistical models summarized from physical-based asymptotic and full-wave simulations. Some models have been verified with measurements. The simulation tool can help save cost and time for the automotive industry, especially for autonomous vehicles. The simulation tool can also help develop new functionalities like target identification or classification as well as help prevent false alarms.