Sensor performance monitoring for autonomous systems and applications
In various examples, a technique for monitoring sensor performance is disclosed that includes aggregating sensor data collected by a sensor into a plurality of statistics for a plurality of sectors corresponding to a two-dimensional (2D) plane representing a sensory field of the sensor. The techniques also include determining a performance of the sensor based on the plurality of statistics and historical sensor data or information associated with the sensor. The techniques further include transmitting, to one or more components, one or more indications of the performance of the sensor to cause the one or more components to perform one or more operations in view of or accounting for the performance of the sensor.
1 . A method, comprising:
aggregating sensor data collected using a sensor into a plurality of statistics for a plurality of sectors of a two-dimensional (2D) plane corresponding to a sensory field of the sensor, the plurality of statistics being based on statistical consistency between sensor data for a subsector within a sector in the plurality of sectors and sensor data over the sector in the plurality of sectors;
determining a performance level of the sensor based at least on the plurality of statistics and historical information associated with the sensor, the performance level being based at least on posterior probabilities calculated for the plurality of statistics relative to a plurality of hypotheses corresponding to different performance levels for the sensor; and
configuring an autonomous system to perform one or more operations based at least on one or more indications of the performance level of the sensor, the autonomous system to compensate when the one or more indications of the performance level of the sensor indicate that the sensor is not performing as expected.
2 . The method of claim 1 , wherein the aggregating the sensor data comprises performing a first aggregation of the sensor data along a first plurality of sectors into which the 2D plane is divided.
3 . The method of claim 2 , wherein aggregating the sensor data further comprises performing a second aggregation of the sensor data along a second plurality of sectors into which the first plurality of sectors is divided.
4 . The method of claim 3 , wherein the second plurality of sectors comprises at least one of a rectangular region, a triangular region, or a polygonal region having sides with non-uniform lengths.
5 . The method of claim 2 , wherein the first plurality of sectors comprises at least one of a first region that spans a width of the 2D plane or a second region that spans a height of the 2D plane.
6 . The method of claim 1 , wherein the posterior probabilities are calculated based at least on a prior probability associated with the historical information and the plurality of statistics.
7 . The method of claim 6 , wherein the prior probability is determined based at least on a comparison of the historical information and at least one statistic included in the plurality of statistics.
8 . The method of claim 6 , wherein the hypotheses comprise at least one of a full blockage of the sensor, a partial blockage of the sensor, or a range of the sensor.
9 . The method of claim 1 , wherein the one or more indications of the performance level of the sensor comprise an aggregation of one or more posterior probabilities associated with the performance level of the sensor across the 2D plane.
10 . The method of claim 1 , wherein the historical information comprises at least one of a previously generated plurality of statistics associated with the sensor or one or more sets of sensor data associated with a sensor type of the sensor.
11 . A processor comprising:
one or more processing units to perform operations comprising:
aggregating sensor data collected using a sensor into a plurality of statistics corresponding to a plurality of sectors of a tessellated two-dimensional (2D) plane associated with the sensor, the plurality of statistics being based on statistical consistency between sensor data for a subsector within a sector in the plurality of sectors and sensor data over the sector in the plurality of sectors;
determining one or more performance values corresponding to the sensor based at least on the plurality of statistics and historical information associated with the sensor, the one or more performance values being based at least on posterior probabilities calculated for the plurality of statistics relative to a plurality of hypotheses corresponding to different performance levels for the sensor; and
configuring an autonomous system to perform one or more operations based at least on one or more indications of the one or more performance values, the autonomous system to compensate when the one or more indications of the one or more performance values indicate that the sensor is not performing as expected.
12 . The processor of claim 11 , wherein the aggregating the sensor data comprises:
performing a first aggregation of the sensor data along a first plurality of sectors of the tessellated 2D plane; and
performing a second aggregation of the sensor data along a second plurality of sectors into which the first plurality of sectors is divided.
13 . The processor of claim 11 , wherein the operations further comprise tessellating a 2D plane corresponding to a sensory field of the sensor to generate the tessellated 2D plane.
14 . The processor of claim 11 , wherein the posterior probabilities are calculated based at least on a prior probability associated with the historical information and the plurality of statistics.
15 . The processor of claim 14 , wherein the posterior probabilities are determined based at least on a first statistic for a first sector and second statistic for a second sector that corresponds to a subset of the first sector.
16 . The processor of claim 11 , wherein the plurality of statistics comprises at least one of an average, a sum, a ratio, a histogram, or an integration.
17 . The processor of claim 11 , wherein the sensor comprises at least one of a camera, a LiDAR sensor, a RADAR sensor, or an ultrasonic sensor.
18 . The processor of claim 11 , wherein the processor is comprised in at least one of:
a control system for an autonomous or semi-autonomous machine;
a perception system for an autonomous or semi-autonomous machine;
a system for performing simulation operations;
a system for performing digital twin operations;
a system for performing light transport simulation;
a system for performing collaborative content creation for 3D assets;
a system for performing deep learning operations;
a system implemented using an edge device;
a system for generating or presenting at least one of virtual reality content, augmented reality content, or mixed reality content;
a system implemented using a robot;
a system for performing conversational AI operations;
a system for generating synthetic data;
a system incorporating one or more virtual machines (VMs);
a system implemented at least partially in a data center; or
a system implemented at least partially using cloud computing resources.
19 . A system comprising:
one or more processing units to perform one or more autonomous operations associated with an autonomous system based at least on an indication of a current operating condition of a sensor, the current operating condition of the sensor determined based at least on comparing historical sensor information to a plurality of statistics corresponding to a plurality of sectors associated with a two-dimensional (2D) tessellation of a sensory field of the sensor and statistical consistency between sensor data for a subsector within a sector in the plurality of sectors and sensor data over the sector in the plurality of sectors, the current operating condition determined based on posterior probabilities calculated for the plurality of statistics relative to a plurality of hypotheses corresponding to different performance levels for the sensor, the one or more processing units to compensate when the indication of the current operating condition of the sensor indicates that the sensor is not performing as expected.
20 . The system of claim 19 , wherein the system is comprised in at least one of:
a control system for an autonomous or semi-autonomous machine;
a perception system for an autonomous or semi-autonomous machine;
a system for performing simulation operations;
a system for performing digital twin operations;
a system for performing light transport simulation;
a system for performing collaborative content creation for 3D assets;
a system for performing deep learning operations;
a system implemented using an edge device;
a system for generating or presenting at least one of virtual reality content, augmented reality content, or mixed reality content;
a system implemented using a robot;
a system for performing conversational AI operations;
a system for generating synthetic data;
a system incorporating one or more virtual machines (VMs);
a system implemented at least partially in a data center; or
a system implemented at least partially using cloud computing resources.