Precipitation rate determination for autonomous vehicles
View Patent ↗Disclosed herein are system, method, and computer program product embodiments for determining a precipitation rate. In some embodiments, a processor receives precipitation data from an array of sensors of a set of array of sensors disposed on a vehicle. A set of lidar sensors are disposed on the vehicle and the array of sensors is positioned within a predetermined distance of a respective lidar sensor of the set of lidar sensors. The processor generates a precipitation rate based on an average of the precipitation data received from the array of sensors and trains a model to control a function of the vehicle during precipitation using the precipitation rate.
1 . A computer-implemented method, comprising:
generating a precipitation rate based on an average of precipitation data from an array of a plurality of precipitation related sensors disposed on a vehicle according to a topology and positioned so that a center of the topology is within a predetermined distance of a respective lidar sensor of a set of lidar sensors disposed on the vehicle, wherein a distance between two different lidar sensors of the set of lidar sensors is larger than a distance between the center of the topology to the respective lidar sensor;
generating a training dataset based on the precipitation rate; and
training a machine learning model configured to control a function of the vehicle using the training dataset.
2 . The computer-implemented method of claim 1 , further comprising validating data collected using the set of lidar sensors based on the precipitation rate.
3 . The computer-implemented method of claim 1 , wherein the machine learning model and the function of the vehicle correspond to a position of the array of sensors on the vehicle.
4 . The computer-implemented method of claim 1 , wherein the training dataset is further based a droplet size of precipitation determined based on the precipitation data.
5 . The computer-implemented method of claim 1 , further comprising collecting the precipitation data while the vehicle is in-motion.
6 . The computer-implemented method of claim 1 , further comprising collecting the precipitation data while the vehicle is in a wind tunnel.
7 . The computer-implemented method of claim 1 , further comprising detecting precipitation that interrupts an infrared beam bounced off of a lens.
8 . A system, the system comprising:
a memory; and
a processor coupled to the memory, wherein the processor is configured to:
generate a precipitation rate based on an average of precipitation data from an array of a plurality of precipitation related sensors disposed on a vehicle according to a topology and positioned so that a center of the topology is within a predetermined distance of a respective lidar sensor of a set of lidar sensors disposed on the vehicle, wherein a distance between two different lidar sensors of the set of lidar sensors is larger than a distance between the center of the topology to the respective lidar sensor;
generate a training dataset based on the precipitation rate; and
train a machine learning model to control a function of the vehicle during precipitation using the precipitation rate.
9 . The system of claim 8 , wherein the precipitation rate is used to validate data collected using the set of lidar sensors.
10 . The system of claim 8 , wherein the machine learning model and the function correspond to a position of the array of sensors on the vehicle.
11 . The system of claim 8 , wherein the processor is further configured to:
determine a droplet size of precipitation based on the precipitation data; and
train the machine learning model to control the function of the vehicle during the precipitation using the droplet size.
12 . The system of claim 8 , wherein the vehicle is in-motion while the array of sensors collects the precipitation data.
13 . The system of claim 8 , wherein the vehicle is in a wind tunnel while the array of sensors collects the precipitation data.
14 . The system of claim 8 , wherein a sensor of the array of sensors bounces an infrared beam off of a lens and the sensor is configured to detect precipitation that interrupts the infrared beam.
15 . A non-transitory computer-readable medium having instructions stored thereon that, when executed by at least one computing device, cause the at least one computing device to perform operations, the operations comprising:
generating a precipitation rate based on an average of precipitation data from an array of a plurality of precipitation related sensors disposed on a vehicle according to a topology and positioned so that a center of the topology is within a predetermined distance of a respective lidar sensor of a set of lidar sensors disposed on the vehicle, wherein a distance between two different lidar sensors of the set of lidar sensors is larger than a distance between the center of the topology to the respective lidar sensor;
generating a training dataset based on the precipitation rate; and
training a model to control a function of the vehicle during precipitation using the precipitation rate.
16 . The non-transitory computer-readable medium of claim 15 , wherein the precipitation rate is used to validate data collected using the set of lidar sensors.
17 . The non-transitory computer-readable medium of claim 15 , wherein the model and the function correspond to a position of the array of sensors on the vehicle.
18 . The non-transitory computer-readable medium of claim 15 , wherein the operations further comprise:
determining a droplet size of precipitation based on the precipitation data; and
training the model to control the function of the vehicle during the precipitation using the droplet size.
19 . The non-transitory computer-readable medium of claim 15 , wherein the vehicle is in-motion while the array of sensors collects the precipitation data.
20 . The non-transitory computer-readable medium of claim 15 , wherein the vehicle is in a wind tunnel while the array of sensors collects the precipitation data.