Continuous radar calibration check
Methods and apparatus consistent with the present disclosure may compare position data associated with a Light Detection and Ranging (LiDAR) device and location data associated with a radar device. Changes in relative locations of objects detected along a roadway over time may be used to identify whether sensing capabilities of a sensing apparatus are acceptable. Changes in positions detected by operation of a LiDAR device and locations detected by operation of a radar device may be considered normal or acceptable when those changes are below a threshold value. In instances when changes in relative positions/locations detected by operation the LiDAR device and the radar device at an AV change beyond a threshold value, a sensing system may be considered as being out of calibration. Once a processor detects that a sensing system has changed close to or beyond a threshold level, that processor may initiate a corrective action.
1. A method for monitoring a sensing apparatus, the method comprising:
identifying a position of an object based on data associated with a light detecting and ranging (LiDAR) sensor, wherein LiDAR sensor data corresponds to a current operational mode of the sensing apparatus;
identifying a location of the object based on data associated with a radar sensor, wherein radar sensor data corresponds to the current operational mode of the sensing apparatus;
identifying a point distance to associate the position and the location;
identifying that the point distance is less than a threshold distance based on comparing the position to the location;
continuing operation of the LiDAR sensor and the radar sensor based on the identification that the point distance is less than the threshold distance;
receiving LiDAR calibration data based on light reflecting off of a target used to initially calibrate the sensing apparatus;
receiving radar calibration data based on a radar signal reflecting off the target when the sensing apparatus is initially calibrated;
associating a first position of the target identified based on an evaluation of the LiDAR calibration data;
associating a first location of the target identified based on an evaluation of the radar calibration data, wherein the first position and the first location correspond to a distance that the LiDAR sensor is separated from the radar sensor at an automated vehicle (AV); and
identifying an offset to associate with at least one of the LiDAR calibration data or the radar calibration data.
2. The method of claim 1 , further comprising:
identifying a second position of a second object based on additional data received via the LiDAR sensor;
identifying a second location of the second object based on additional data associated with the radar sensor;
identifying a second point distance to associate with the second position and the second location; and
issuing a warning indicating that the current operational mode of the sensing apparatus should not be maintained based on an identification that the second point distance is not within the threshold distance.
3. The method of claim 1 , further comprising:
identifying a first set of points of the LiDAR sensor data to associate with the object and with a calculation directed to identifying the point distance;
identifying a second set of points of the radar sensor data to associate with the object and with the calculation directed to identifying the point distance;
pairing respective points from the first set of points with respective points from the second set of points into a number of paired points;
identifying differences to associate with each of the respectively paired points; and
calculating an average by summing the differences associated with each of the respectively paired points and by dividing the sum of the differences by the number of paired points.
4. The method of claim 3 , further comprising:
selecting from the first set of points a number of points of the LiDAR sensor data to pair with the points of the second set of points;
selecting from the second set of points a number of points of the radar sensor data to pair with the points of the first set of points, wherein the number of points of the LiDAR sensor data are equal to the number of points of the radar sensor data; and
performing a calculation to identify the point distance.
5. The method of claim 1 , further comprising:
identifying a first distance associated with the position of the object;
identifying a second distance associated with the location of the object;
identifying a difference between the first distance and the second distance; and
subtracting the offset from the difference between the first distance and the second distance when identifying the point distance.
6. A non-transitory computer-readable storage medium having embodied thereon a program executed by a processor to perform a method for monitoring a sensing apparatus, the method comprising:
identifying a position of an object based on data associated with a light detecting and ranging (LiDAR) sensor, wherein LiDAR sensor data corresponds to a current operational mode of the sensing apparatus;
identifying a location of the object based on data associated with a radar sensor, wherein radar sensor data corresponds to the current operational mode of the sensing apparatus;
identifying a point distance to associate the position and the location;
identifying that the point distance is less than a threshold distance based on comparing the position to the location;
continuing operation of the LiDAR sensor and the radar sensor based on the identification that the point distance is less than the threshold distance;
receive LiDAR calibration data based on light reflecting off of a target used to initially calibrate the sensing apparatus;
receive radar calibration data based on a radar signal reflecting off the target when the sensing apparatus is initially calibrated;
associate a first position of the target identified based on an evaluation of the LiDAR calibration data;
associate a first location of the target identified based on an evaluation of the radar calibration data, wherein the first position and the first location correspond to a distance that the LiDAR sensor is separated from the radar sensor at an automated vehicle (AV); and
identify an offset to associate with at least one of the LiDAR calibration data or the radar calibration data.
7. The non-transitory computer-readable storage medium of claim 6 , further comprising:
identifying a second position of a second object based on additional data received via the LiDAR sensor;
identifying a second location of the second object based on additional data associated with the radar sensor;
identifying a second point distance to associate with the second position and the second location; and
issuing a warning indicating that the current operational mode of the sensing apparatus should not be maintained based on an identification that the second point distance is not within the threshold distance.
8. The non-transitory computer-readable storage medium of claim 6 , the program further executable to:
identify a first set of points of the LiDAR sensor data to associate with the object and with a calculation directed to identifying the point distance;
identify a second set of points of the radar sensor data to associate with the object and with the calculation directed to identifying the point distance;
pair respective points from the first set of points with respective points from the second set of points into a number of paired points;
identify differences to associate with each of the respectively paired points; and
calculate an average by summing the differences associated with each of the respectively paired points and by dividing the sum of the differences by the number of paired points.
9. The non-transitory computer-readable storage medium of claim 8 , the program further executable to:
select from the first set of points a number of points of the LiDAR sensor data to pair with the points of the second set of points;
select from the second set of points a number of points of the radar sensor data to pair with the points of the first set of points, wherein the number of points of the LiDAR sensor data are equal to the number of points of the radar sensor data; and
perform a calculation to identify the point distance.
10. The non-transitory computer-readable storage medium of claim 6 , the program further executable to:
identify a first distance associated with the position of the object; identify a second distance associated with the location of the object;
identify a difference between the first distance and the second distance; and
subtract the offset from the difference between the first distance and the second distance when identifying the point distance.
11. A sensing apparatus comprising:
a light detecting and ranging (LiDAR) sensor that collects LiDAR sensor data from which a position of an object is identified;
a radar sensor that collects radar sensor data from which a location of the object is identified;
a memory; and
a processor that executes instructions out of the memory to:
identify a point distance to associate the position and the location,
identify that the point distance is less than a threshold distance based on comparing the position to the location;
continue operation of the LiDAR sensor and the radar sensor based on an identification that the point distance is less than the threshold distance;
receive LiDAR calibration data based on light reflecting off of a target used to initially calibrate the sensing apparatus;
receive radar calibration data based on a radar signal reflecting off the target when the sensing apparatus is initially calibrated;
associate a first position of the target identified based on an evaluation of the LiDAR calibration data;
associate a first location of the target identified based on an evaluation of the radar calibration data, wherein the first position and the first location correspond to a distance that the LiDAR sensor is separated from the radar sensor at an automated vehicle (AV); and
identify an offset to associate with at least one of the LiDAR calibration data or the radar calibration data.
12. The sensing apparatus of claim 11 , wherein the processor executes the instructions out of the memory to:
identify a second position of a second object based on additional data received via the LiDAR sensor;
identify a second location of the second object based on additional data associated with the radar sensor; identify a second point distance to associate with the second position and the second location; and issue a warning indicating that a current mode of the sensing apparatus should not be maintained based on an identification that the second point distance is not within the threshold distance.
13. The sensing apparatus of claim 11 , wherein the processor executes the instructions out of the memory to:
identify a first set of points of the LiDAR sensor data to associate with the object and with a calculation directed to identifying the point distance;
identify a second set of points of the radar sensor data to associate with the object and with the calculation directed to identifying the point distance;
pair respective points from the first set of points with respective points from the second set of points into a number of paired points;
identify differences to associate with each of the respectively paired points; and
calculate an average by summing the differences associated with each of the respectively paired points and by dividing the sum of the differences by the number of paired points.
14. The sensing apparatus of claim 13 , wherein the processor executes the instructions out of the memory to:
select from the first set of points a number of points of the LiDAR sensor data to pair with the points of the second set of points;
select from the second set of points a number of points of the radar sensor data to pair with the points of the first set of points, wherein the number of points of the LiDAR sensor data are equal to the number of points of the radar sensor data; and
perform a calculation to identify the point distance.