Detection and localization of non-line-of-sight objects using multipath radar reflections and map data
This document describes techniques and systems to detect and localize NLOS objects using multipath radar reflections and map data. In some examples, a processor of radar system can identify a detection of an object using reflected EM energy and determine, using map data, whether a direct-path reflection associated with the detection is within a roadway. In response to determining that the direct-path reflection is not located within the roadway, the processor can determine whether a multipath reflection (e.g., a multipath range and multipath angle) associated with the detection is viable. In response to determining that the multipath reflection is viable, the processor can determine that the detection corresponds to an NLOS object. The processor can also provide the NLOS object as an input to an autonomous or semi-autonomous driving system of the vehicle, thereby improving the safety of such systems.
1. A radar system installed in a host vehicle, the radar system comprising:
one or more processors configured to:
identify, using electromagnetic (EM) energy reflected by one or more objects and received by a receiver of the radar system, a detection of an object;
determine, using the received EM energy and assuming a direct-path reflection, a direct-path azimuth angle and a direct-path range associated with the detection;
determine, using map data, the direct-path azimuth angle, and the direct-path range, whether the direct-path reflection associated with the detection is located within a roadway corresponding to a travel path of the host vehicle;
responsive to a determination that the direct-path reflection associated with the detection is not located within the roadway, determine, using the received EM energy and assuming a multipath reflection, a multipath azimuth angle, a multipath range, and a multipath cross-range associated with the detection;
determine whether the multipath reflection associated with the detection is located within the roadway and not within a direct line-of-sight of the radar system by identifying, based on the map data, whether the multipath range, the multipath cross-range, and the multipath azimuth angle are located within the roadway and not within the direct line-of-sight of the radar system;
responsive to a determination that the multipath reflection associated with the detection is located within the roadway and not within the direct line-of-sight of the radar system, determine that the detection corresponds to a non-line-of-sight (NLOS) object; and
provide an indication of the NLOS object, including the multipath range, the multipath cross-range, and the multipath azimuth angle, as an input to an autonomous-driving system or an assisted-driving system.
2. The radar system of claim 1 , wherein a determination that the multipath reflection associated with the detection is located within the roadway comprises determining whether a multipath position associated with the detection is located within a lane of the roadway.
3. The radar system of claim 1 , wherein the one or more processors are configured to determine whether the multipath reflection associated with the detection is located within the roadway by:
determining whether a reflective object is located along the direct-path azimuth angle associated with the detection;
responsive to determining that the reflective object is located along the direct-path azimuth angle associated with the detection, determining, based on the direct-path azimuth angle associated with the detection, an angle of reflection for the received EM energy at the reflective object;
determining, based on the angle of reflection for the received EM energy at the reflective object, an angle of incidence for the received EM energy at the reflective object; and
determining the multipath range, the multipath cross-range, and multipath azimuth angle associated with the detection based on the angle of incidence for the received EM energy at the reflective object and the direct-path range associated with the detection.
4. The radar system of claim 3 , wherein the one or more processors are configured to determine whether the reflective object is located along the direct-path azimuth angle associated with the detection based on reflective objects included in the map data or reflective objects detected by the radar system, a lidar system, or a vision system.
5. The radar system of claim 1 , wherein:
the NLOS object is moving; and
the one or more processors are further configured to start or update a track associated with the NLOS object.
6. The radar system of claim 5 , wherein the one or more processors are further configured to provide the track associated with the NLOS object as another input to the autonomous-driving system or the assisted-driving system.
7. The radar system of claim 1 , wherein:
the NLOS object is stationary; and
the one or more processors are further configured to start or update a track associated with the NLOS object.
8. The radar system of claim 7 , wherein the one or more processors are further configured to provide the track associated with the NLOS object as another input to the autonomous-driving system or the assisted-driving system.
9. The radar system of claim 1 , wherein the map data is included in a memory of the host vehicle.
10. The radar system of claim 1 , wherein the map data is obtained from a remote computer system.
11. A computer-readable storage medium comprising computer-executable instructions that, when executed, cause a processor of a radar system installed in a host vehicle to:
identify, using electromagnetic (EM) energy reflected by one or more objects and received by a receiver of the radar system, a detection of an object;
determine, using the received EM energy and assuming a direct-path reflection, a direct-path azimuth angle and a direct-path range associated with the detection;
determine, using map data, the direct-path azimuth angle, and the direct-path range, whether the direct-path reflection associated with the detection is located within a roadway corresponding to a travel path of the host vehicle;
responsive to a determination that the direct-path reflection associated with the detection is not located within the roadway, determine, using the received EM energy and assuming a multipath reflection, a multipath azimuth angle, a multipath range, and a multipath cross-range associated with the detection;
determine whether the multipath reflection associated with the detection is located within the roadway and not within a direct line-of-sight of the radar system by identifying, based on the map data, whether the multipath range, the multipath cross-range, and the multipath azimuth angle are located within the roadway and not within the direct line-of-sight of the radar system;
responsive to a determination that the multipath reflection associated with the detection is located within the roadway and not within the direct line-of-sight of the radar system, determine that the detection corresponds to a non-line-of-sight (NLOS) object; and
provide an indication of the NLOS object, including the multipath range, the multipath cross-range, and the multipath azimuth angle, as an input to an autonomous-driving system or an assisted-driving system.
12. The computer-readable storage medium of claim 11 , wherein the computer-readable storage medium further comprises computer-executable instructions that, when executed, cause the processor of the radar system to determine whether the multipath reflection associated with the detection is located within the roadway by:
determining whether a reflective object is located along the direct-path azimuth angle associated with the detection;
responsive to determining that a reflective object is located along the direct-path azimuth angle associated with the detection, determining, based on the direct-path azimuth angle associated with the detection, an angle of reflection for the received EM energy at the reflective object;
determining, based on the angle of reflection for the received EM energy at the reflective object, an angle of incidence for the received EM energy at the reflective object; and
determining the multipath range, the multipath cross-range, and the multipath azimuth angle associated with the detection based on the angle of incidence for the received EM energy at the reflective object and the direct-path range associated with the detection.
13. A method comprising:
identifying, using electromagnetic (EM) energy reflected by one or more objects and received by a receiver of a radar system installed in a host vehicle, a detection of an object;
determining, using the received EM energy and assuming a direct-path reflection, a direct-path azimuth angle and a direct-path range associated with the detection;
determining, using map data, the direct-path azimuth angle, and the direct-path range, whether the direct-path reflection associated with the detection is located within a roadway corresponding to a travel path of the host vehicle;
responsive to a determination that the direct-path reflection associated with the detection is not located within the roadway, determining, using the received EM energy and assuming a multipath reflection, a multipath azimuth angle, a multipath range, and a multipath cross-range associated with the detection;
determining whether the multipath reflection associated with the detection is located within the roadway and not within a direct line-of-sight of the radar system by identifying, based on the map data, whether the multipath range, the multipath cross-range, and the multipath azimuth angle are located within the roadway and not within the direct line-of-sight of the radar system;
responsive to a determination that the multipath reflection associated with the detection is located within the roadway and not within the direct line-of-sight of the radar system, determining that the detection corresponds to a non-line-of-sight (NLOS) object; and
providing an indication of the NLOS object, including the multipath range, the multipath cross-range, and the multipath azimuth angle, as an input to an autonomous-driving system or an assisted-driving system.