Adjustable sensor housing
The disclosed technology provides solutions for repositioning an adjustable sensor with respect to the autonomous vehicle in response to a determination that the field of view of the sensor is obstructed. A method of the disclosed technology can include steps for actuating an adjustable sensor housing comprising receiving a first set of sensor data representing an environment around an autonomous vehicle (AV), wherein the first set of sensor data is collected using a sensor mounted to the AV via an adjustable sensor housing, determining, based on the first set of sensor data, that a field of view (FOV) for the sensor is obstructed, and in response to the determination that the FOV for the sensor is obstructed, actuating the adjustable sensor housing to reposition the sensor with respect to the AV. Systems and machine-readable media are also provided.
1 . An apparatus comprising:
at least one memory; and
at least one processor coupled to the at least one memory, the at least one processor configured to:
receive a first set of sensor data representing an environment around an autonomous vehicle (AV), wherein the first set of sensor data is collected using a sensor mounted to the AV via an adjustable sensor housing;
determine, based on the first set of sensor data and map data, that a field of view (FOV) for a second sensor is obstructed;
determine, based on the first set of sensor data and the map data, that the obstruction is permanent;
based on the determination that the obstruction is permanent, update the map data to include data indicating a location of the permanent obstruction and a predetermined repositioning location for the second sensor associated with the permanent obstruction; and
based on the determination that the FOV for the second sensor is obstructed, actuate the adjustable sensor housing to reposition the second sensor with respect to the AV to the predetermined repositioning location indicated in the updated map data.
2 . The apparatus of claim 1 , wherein to determine that the FOV for the second sensor is obstructed, the at least one processor is further configured to provide the first set of sensor data to a perception layer of a software stack for the AV;
identify a semantic label for at least one object in the FOV for the second sensor; and
determine that the at least one object is occluded within the FOV based on the semantic label.
3 . The apparatus of claim 1 , wherein the at least one processor is further configured to:
receive a second set of sensor data; and
determine, based on the second set of sensor data, if the FOV for the second sensor is obstructed.
4 . The apparatus of claim 1 , wherein to actuate the adjustable sensor housing to reposition the second sensor with respect to the AV, the at least one processor is further configured to:
actuate one or more motors disposed in the adjustable sensor housing to modify a pitch of the second sensor.
5 . The apparatus of claim 1 , wherein to actuate the adjustable sensor housing to reposition the second sensor with respect to the AV, the at least one processor is further configured to:
actuate one or more motors disposed in the adjustable sensor housing to modify a roll of the second sensor.
6 . The apparatus of claim 1 , wherein to actuate the adjustable sensor housing to reposition the second sensor with respect to the AV, the at least one processor is further configured to:
actuate one or more motors disposed in the adjustable sensor housing to modify a yaw of the second sensor.
7 . The apparatus of claim 1 , wherein the second sensor is a Light Detection and Ranging (LiDAR) sensor, a camera sensor, or a Radio Detection and Ranging (RADAR) sensor.
8 . A computer-implemented method for actuating an adjustable sensor housing comprising:
receiving a first set of sensor data representing an environment around an autonomous vehicle (AV), wherein the first set of sensor data is collected using a sensor mounted to the AV via the adjustable sensor housing;
determining, based on the first set of sensor data and map data, that a field of view (FOV) for a second sensor is obstructed;
determining, based on the first set of sensor data and the map data, that the obstruction is permanent;
based on the determination that the obstruction is permanent, updating the map data to include data indicating a location of the permanent obstruction and a predetermined repositioning location for the second sensor associated with the permanent obstruction; and
based on the determination that the FOV for the second sensor is obstructed, actuating the adjustable sensor housing to reposition the second sensor with respect to the AV to the predetermined repositioning location indicated in the updated map data.
9 . The computer-implemented method of claim 8 , wherein determining that the FOV for the second sensor is obstructed, further comprises:
providing the first set of sensor data to a perception layer of a software stack for the AV;
identifying a semantic label for at least one object in the FOV for the second sensor; and
determining that the at least one object is occluded within the FOV based on the semantic label.
10 . The computer-implemented method of claim 8 , further comprising:
receiving a second set of sensor data; and
determining, based on the second set of sensor data, if the FOV for the second sensor is obstructed.
11 . The computer-implemented method of claim 8 , wherein actuating the adjustable sensor housing to reposition the second sensor with respect to the AV, further comprises:
actuating one or more motors disposed in the adjustable sensor housing to modify a pitch of the second sensor.
12 . The computer-implemented method of claim 8 , wherein actuating the adjustable sensor housing to reposition the second sensor with respect to the AV, further comprises:
actuating one or more motors disposed in the adjustable sensor housing to modify a roll of the second sensor.
13 . The computer-implemented method of claim 8 , wherein actuating the adjustable sensor housing to reposition the second sensor with respect to the AV, further comprises:
actuating one or more motors disposed in the adjustable sensor housing to modify a yaw of the second sensor.
14 . The computer-implemented method of claim 8 , wherein the second sensor is a Light Detection and Ranging (LiDAR) sensor, a camera sensor, or a Radio Detection and Ranging (RADAR) sensor.
15 . A non-transitory computer-readable storage medium comprising at least one instruction for causing a computer or processor to:
receive a first set of sensor data representing an environment around an autonomous vehicle (AV), wherein the first set of sensor data is collected using a sensor mounted to the AV via an adjustable sensor housing;
determine, based on the first set of sensor data and map data, that a field of view (FOV) for a second sensor is obstructed;
determine, based on the first set of sensor data and the map data, that the obstruction is permanent;
based on the determination that the obstruction is permanent, update the map data to include data indicating a location of the permanent obstruction and a predetermined repositioning location for the second sensor associated with the permanent obstruction; and
based on the determination that the FOV for the second sensor is obstructed, actuate the adjustable sensor housing to reposition the second sensor with respect to the AV to the predetermined repositioning location indicated in the updated map data.
16 . The non-transitory computer-readable storage medium of claim 15 , wherein to determine that the FOV for the second sensor is obstructed, the at least one instruction causes the computer or processor to:
provide the first set of sensor data to a perception layer of a software stack for the AV;
identify a semantic label for at least one object in the FOV for the second sensor; and
determine that the at least one object is occluded within the FOV based on the semantic label.
17 . The non-transitory computer-readable storage medium of claim 15 , wherein the at least one instruction causes the computer or processor to:
receive a second set of sensor data; and
determine, based on the second set of sensor data, if the FOV for the second sensor is obstructed.
18 . The non-transitory computer-readable storage medium of claim 15 , wherein to actuate the adjustable sensor housing to reposition the second sensor with respect to the AV, the at least one instruction causes the computer or processor to:
actuate one or more motors disposed in the adjustable sensor housing to modify a pitch of the second sensor.
19 . The non-transitory computer-readable storage medium of claim 15 , wherein to actuate the adjustable sensor housing to reposition the second sensor with respect to the AV, the at least one instruction causes the computer or processor to:
actuate one or more motors disposed in the adjustable sensor housing to modify a roll of the second sensor.
20 . The non-transitory computer-readable storage medium of claim 15 , wherein to actuate the adjustable sensor housing to reposition the second sensor with respect to the AV, the at least one instruction causes the computer or processor to:
actuate one or more motors disposed in the adjustable sensor housing to modify a yaw of the second sensor.