Systems and methods for using piezoelectric sensors to detect alignment anomaly

Abstract

Systems and methods are provided for detecting an enclosure alignment anomaly. Pressure data of a set period can be obtained from one or more piezoelectric sensors. The one or more piezoelectric sensors are installed in between an enclosure and a fixture of an autonomous vehicle. The pressure data of the set period can be processed over a period of time. One or more trends can be identified based on the processed pressure data.

1 . A method for detecting an enclosure alignment anomaly comprising:
 obtaining pressure data from one or more piezoelectric sensors, the one or more piezoelectric sensors being installed in between an enclosure and a fixture mounted to the enclosure;   processing the pressure data; and   identifying one or more trends based on the processed pressure data.   
     
     
         2 . The method of  claim 1 , wherein the pressure data is over a set period. 
     
     
         3 . The method of  claim 2 , wherein the set period is at least one of hourly, daily, weekly, bi-weekly, or monthly. 
     
     
         4 . The method of  claim 1 , wherein processing the pressure data comprises:
 aggregating the pressure data; and   identifying a maximum pressure, a minimum pressure, and an average pressure corresponding to the pressure data.   
     
     
         5 . The method of  claim 4 , wherein processing the pressure data further comprises:
 trending the pressure data; and   determining a nominal range for the pressure data, the nominal range determined based on identifying an upper bound and a lower bound of the pressure data.   
     
     
         6 . The method of  claim 5 , wherein the upper bound is determined by identifying a highest value in the pressure data, and the lower bound is determined by identifying a lowest value in the pressure data. 
     
     
         7 . The method of  claim 1 , wherein identifying the one or more trends based on the processed pressure data comprises:
 identifying a pressure data point in the pressure data that falls outside of a nominal range; and   identifying the pressure data point as an enclosure alignment anomaly.   
     
     
         8 . The method of  claim 1 , wherein identifying the one or more trends based on the processed pressure data comprises:
 trending an average pressure based on the pressure data;   determining a trend based on the trending of the average pressure using at least a regression technique; and   identifying the trend as a potential premature enclosure alignment anomaly.   
     
     
         9 . The method of  claim 1 , wherein identifying one or more trends based on the processed pressure data comprises:
 training a machine learning model using a training data set;   receiving the processed pressure data; and   determining, based on the processed pressure data, an existence of a potential premature enclosure alignment anomaly.   
     
     
         10 . The method of  claim 9 , wherein the machine learning model is implemented using at least one of a classifier or a neural network, and the training data set is based on a portion of the processed pressure data with human annotations. 
     
     
         11 . A system for detecting an enclosure alignment anomaly comprising:
 an enclosure mounted to a fixture;   one or more piezoelectric sensors installed in between the enclosure and the fixture; and   one or more processors; and   a memory storing instructions that, when executed by the one or more processors, cause the one or more processors to perform:
 obtaining pressure data from the one or more piezoelectric sensors; 
 processing the pressure data; and 
 identify one or more trends based on the processed pressure data. 
   
     
     
         12 . The system of  claim 11 , wherein the pressure data is over a set period. 
     
     
         13 . The system of  claim 12 , wherein the set period is at least one of hourly, daily, weekly, bi-weekly, or monthly. 
     
     
         14 . The system of  claim 11 , wherein processing the pressure data comprises:
 aggregating the pressure data; and   identifying a maximum pressure, a minimum pressure, and an average pressure corresponding to the pressure data.   
     
     
         15 . The system of  claim 14 , wherein processing the pressure data further comprises:
 trending the pressure data; and   determining a nominal range for the pressure data, the nominal range determined based on identifying an upper bound and a lower bound of the pressure data.   
     
     
         16 . The system of  claim 15 , wherein the upper bound is determined by identifying a highest value in the pressure data, and the lower bound is determined by identifying a lowest value in the pressure data. 
     
     
         17 . The system of  claim 11 , wherein identifying the one or more trends based on the processed pressure data comprises:
 identifying a pressure data point in the pressure data that falls outside of a nominal range; and   identifying the pressure data point as an enclosure alignment anomaly.   
     
     
         18 . The system of  claim 11 , wherein identifying the one or more trends based on the processed pressure data comprises:
 trending an average pressure based on the pressure data;   determining a trend based on the trending of the average pressure using at least a regression technique; and   identifying the trend as a potential premature enclosure alignment anomaly.   
     
     
         19 . The system of  claim 11 , wherein identifying one or more trends based on the processed pressure data comprises:
 training a machine learning model using a training data set;   receiving the processed pressure data; and   determining, based on the processed pressure data, an existence of a potential premature enclosure alignment anomaly.   
     
     
         20 . The system of  claim 19 , wherein the machine learning model is implemented using at least one of a classifier or a neural network, and the training data set is based on a portion of the processed pressure data with human annotations.