Search

US-12620163-B2 - High-fidelity micro-doppler effect simulator

US12620163B2US 12620163 B2US12620163 B2US 12620163B2US-12620163-B2

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.

Inventors

  • Xiuzhang Cai
  • Bruno F. Camps Raga
  • James Searcy

Assignees

  • Aptiv Technologies AG

Dates

Publication Date
20260505
Application Date
20240109

Claims (20)

  1. 1 . A computer-implemented method comprising: receiving environment description data associated with a simulated environment including an object; identifying an object type of the object; in response to the object type being a human object type: 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; performing ray tracing simulation including: launching a set of rays from a ray source towards the object; and determining a set of propagation paths of the set of rays from the ray source to the object and a set of reflection paths of the set of rays from the object to a ray receiver; performing physical optics simulation to determine a set of scattered fields associated with the set of reflection paths of the set of rays; determining a set of micro-motions of the object based on calculating at least one of a Doppler shift and a micro-Doppler shift for each ray of the set of rays; performing ray clustering including combining rays of the set of rays that include one or more similar characteristics; and transforming a simulation output for display on a user device; wherein: the simulation output is based on the ray tracing simulation and the physical optics simulation, the simulation output indicates the motion type of the object, and the motion type of the object includes a main motion of the object and the set of micro-motions of the object.
  2. 2 . The computer-implemented method of claim 1 wherein the object type is at least one of a human object type, a building object type, a vehicle object type, an animal object type, and a tree object type.
  3. 3 . The computer-implemented method of claim 1 wherein: the motion type is associated with a physical activity of the object; and the physical activity is associated with one of walking, running, dancing, jumping, and riding a bicycle.
  4. 4 . The computer-implemented method of claim 1 wherein: loading the CAD model associated with the object includes selecting the CAD model from a CAD model database; and loading the motion file associated with the motion type includes selecting the motion file from a motion file database.
  5. 5 . The computer-implemented method of claim 1 wherein identifying the motion type associated with the object includes: determining a speed of the object; in response to the speed being greater than a determined threshold, selecting a running motion type as the motion type; and in response to the speed being less than the determined threshold, selecting a walking motion type as the motion type.
  6. 6 . The computer-implemented method of claim 1 further comprising: modifying a mesh of the object; calculating a speed of each facet of the object; generating an acceleration data structure; and copying the acceleration data structure to a graphics processing unit in connection with performing the ray tracing simulation.
  7. 7 . The computer-implemented method of claim 1 wherein: the ray source and the ray receiver are associated with a reference point; and the reference point is based on a radar system of a vehicle.
  8. 8 . The computer-implemented method of claim 1 wherein mapping the CAD model to the motion file includes: linking vertices of a mesh of the object to a first set of bones of the CAD model; mapping a second set of bones of the motion file to the first set of bones; initializing a transforming matrix; and identifying a period of one in cycle in time and distance of the motion file.
  9. 9 . The computer-implemented method of claim 1 further comprising, in response to the ray receiver receiving a ray of the set of rays that hit the object, querying an object tree data structure to identify the object type of the object.
  10. 10 . The computer-implemented method of claim 1 wherein: in response to the object type being the human object type, the ray tracing simulation is performed in connection with a global coordinate system; and in response to the object type not being the human object type, the ray tracing simulation is performed in connection with a local coordinate system.
  11. 11 . A computer system comprising: memory hardware configured to store instructions; and processor hardware configured to execute the instructions, wherein the instructions include: receiving environment description data associated with a simulated environment including an object; identifying an object type of the object; in response to the object type being a human object type: 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; performing ray tracing simulation including: launching a set of rays from a ray source towards the object; and determining a set of propagation paths of the set of rays from the ray source to the object and a set of reflection paths of the set of rays from the object to a ray receiver; performing physical optics simulation to determine a set of scattered fields associated with the set of reflection paths of the set of rays; determining a set of micro-motions of the object based on calculating at least one of a Doppler shift and a micro-Doppler shift for each ray of the set of rays; performing ray clustering including combining rays of the set of rays that include one or more similar characteristics; and transforming a simulation output for display on a user device; wherein: the simulation output is based on the ray tracing simulation and the physical optics simulation, the simulation output indicates the motion type of the object, and the motion type of the object includes a main motion of the object and a set of micro-motions of the object.
  12. 12 . The computer system of claim 11 wherein the object type is at least one of a human object type, a building object type, a vehicle object type, an animal object type, and a tree object type.
  13. 13 . The computer system of claim 11 wherein: the motion type is associated with a physical activity of the object; and the physical activity is associated with one of walking, running, dancing, jumping, and riding a bicycle.
  14. 14 . The computer system of claim 11 wherein: loading the CAD model associated with the object includes selecting the CAD model from a CAD model database; and loading the motion file associated with the motion type includes selecting the motion file from a motion file database.
  15. 15 . The computer system of claim 11 wherein identifying the motion type associated with the object includes: determining a speed of the object; in response to the speed being greater than a determined threshold, selecting a running motion type as the motion type; and in response to the speed being less than the determined threshold, selecting a walking motion type as the motion type.
  16. 16 . The computer system of claim 11 wherein the instructions include: modifying a mesh of the object; calculating a speed of each facet of the object; generating an acceleration data structure; and copying the acceleration data structure to a graphics processing unit in connection with performing the ray tracing simulation.
  17. 17 . The computer system of claim 11 wherein: the ray source and the ray receiver are associated with a reference point; and the reference point is based on a radar system of a vehicle.
  18. 18 . The computer system of claim 11 wherein mapping the CAD model to the motion file includes: linking vertices of a mesh of the object to a first set of bones of the CAD model; mapping a second set of bones of the motion file to the first set of bones; initializing a transforming matrix; and identifying a period of one in cycle in time and distance of the motion file.
  19. 19 . The computer system of claim 11 wherein the instructions include, in response to the ray receiver receiving a ray of the set of rays that hit the object, querying an object tree data structure to identify the object type of the object.
  20. 20 . The computer system of claim 11 wherein: in response to the object type being the human object type, the ray tracing simulation is performed in connection with a global coordinate system; and in response to the object type not being the human object type, the ray tracing simulation is performed in connection with a local coordinate system.

Description

FIELD The present disclosure relates to electromagnetic simulation and more particularly to simulation of electromagnetic environments for radars in wheeled vehicles. BACKGROUND Radars are useful devices that can detect and track objects. In comparison to other types of sensors, for example cameras, radars provide improved performance in difficult environmental conditions, such as low lighting, fog, and/or with moving or overlapping objects. Accordingly, radars provide many advantages for driver assistance applications and/or autonomous driving applications, among others. While designing and developing a radar, engineers may perform live tests (for example, real-time tests) to evaluate performance of the radar. Live tests can be expensive and time-intensive, and typically occur after design and implementation of a radar. As such, there can be a lot of rework and additional cost associated with addressing problems revealed by a live test. Accordingly, radar simulation via a radar simulator may be performed instead of live tests to evaluate performance of a radar. Radar simulation may facilitate discovery of problems sooner in the development of a radar and may be faster and less expensive in comparison with live tests. SUMMARY 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 object type, 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 including launching a set of rays from a ray source towards the object and determining a set of propagation paths of the set of rays from the ray source to the object and a set of reflection paths of the set of rays from the object to a ray receiver. The method includes performing physical optics simulation to determine a set of scattered fields associated with the set of reflection paths of the set of rays, calculating at least one of a Doppler shift and a micro-Doppler shift for each ray of the set of rays, performing ray clustering including combining rays of the set of rays that include one or more similar characteristics, 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, and the motion simulation includes a main motion of the object and a set of micro-motions of the object. In other features, the object type is at least one of a human object type, a building object type, a vehicle object type, an animal object type, and a tree object type. In other features, the motion type is associated with a physical activity of the object. The physical activity is associated with one of walking, running, dancing, jumping, and riding a bicycle. In other features, loading the CAD model associated with the object includes selecting the CAD model from a CAD model database. In other features, loading the motion file associated with the motion type includes selecting the motion file from a motion file database. In other features, identifying the motion type associated with the object includes determining a speed of the object. In response to the speed being greater than a determined threshold, selecting a running motion type as the motion type. In response to the speed being less than the determined threshold, selecting a walking motion type as the motion type. In other features, the method includes modifying a mesh of the object, calculating a speed of each facet of the object, generating an acceleration data structure, and copying the acceleration data structure to a graphics processing unit in connection with performing the ray tracing simulation. In other features, the ray source and the ray receiver are associated with a reference point. The reference point is based on a radar system of a vehicle. In other features, mapping the CAD model to the motion file includes linking vertices of a mesh of the object to a first set of bones of the CAD model, mapping a second set of bones of the motion file to the first set of bones, initializing a transforming matrix, and identifying a period of one in cycle in time and distance of the motion file. In other features, the method includes, in response to the ray receiver receiving a ray of the set of rays that hit the object, querying an object tree data structure to identify the object type of the object. In other features, in response to the object type being the human object type, the ray tracing simulation is performed in connection with a global coordinate system. In response to the object type not being the human object type, the ray tracing simulation is performed in connection with a local coordinate system. A computer system i