System and method for single-loop vehicle speed estimation using inductive loop signature data

Abstract

A system and method are disclosed for using vehicle signature data from a single inductive loop sensor to estimate vehicle speed. Two regressors are calculated from the vehicle signature: the “inverse of the duration” and the “slew rate”. A point is found by normalizing the vehicle signature in amplitude and determining the earliest point in the normalized signature to cross a set threshold. The slew rate is the slope of the normalized vehicle signature at this point. Regression models are generated from empirical data for several vehicle categories. Using the regression model for the category of vehicle and the two regressors vehicle speed is estimated. The regression models have been demonstrated to be robust eliminating the need for site-specific calibration or estimation.

What is claimed is: 
     
         1 - 7 . (canceled) 
     
     
         8 . A method for estimating a speed of a vehicle, the method comprising:
 receiving a vehicle signature signal from an inductive loop sensor through a hardware interface upon traversal of the vehicle over the inductive loop sensor;   operating a hardware processor to:
 measure a temporal duration of the vehicle signature signal; 
 determine a first regressor from the temporal duration; 
 measure a slope at a particular point of the vehicle signature signal; 
 determine a second regressor from the slope; 
 select a regression model based on a category in which the vehicle is classified; and 
 estimate vehicle speed with the regression model using the first regressor and the second regressor; and 
   recording the vehicle speed to a non-volatile, computer-readable storage medium.   
     
     
         9 . The method of  claim 8 , wherein operating the hardware processor to determine the first regressor comprises calculating an inverse of the temporal duration and storing the inverse of the temporal duration as the first regressor. 
     
     
         10 . The method of  claim 8 , wherein operating the processor to measure the slope comprises operating the processor to:
 normalize the vehicle signature signal in amplitude;   identify a first time point, the first time point being an earliest time point at which the amplitude of the normalized vehicle signature signal exceeds a threshold value; and   calculating the slope of the normalized vehicle signature signal at the first time point.   
     
     
         11 . A method for estimating a speed of a vehicle, the method comprising:
 receiving a vehicle signature signal from an inductive loop sensor through a hardware interface upon traversal of the vehicle over the inductive loop sensor;   operating a hardware processor to:
 measure a temporal duration of the vehicle signature signal; 
 determine a first regressor from the temporal duration; 
 measure a slope at a particular point of the vehicle signature signal; 
 determine a second regressor from the slope; 
 select a regression model based on a category in which the vehicle is classified; and 
 estimate vehicle speed with the regression model using the first regressor and the second regressor; and 
   recording the vehicle speed to a non-volatile, computer-readable storage medium;   wherein operating the processor to measure the slope comprises operating the processor to:
 normalize the vehicle signature signal in amplitude; 
 identify a first time point, the first time point being an earliest time point at which the amplitude of the normalized vehicle signature signal exceeds a threshold value; and 
 calculating the slope of the normalized vehicle signature signal at the first time point; 
   wherein the threshold value is between 0.3 and 0.5.   
     
     
         12 . A method for estimating a speed of a vehicle, the method comprising:
 receiving a vehicle signature signal from an inductive loop sensor through a hardware interface upon traversal of the vehicle over the inductive loop sensor;   operating a hardware processor to:
 measure a temporal duration of the vehicle signature signal; 
 determine a first regressor from the temporal duration; 
 measure a slope at a particular point of the vehicle signature signal; 
 determine a second regressor from the slope; 
 select a regression model based on a category in which the vehicle is classified; and 
 estimate vehicle speed with the regression model using the first regressor and the second regressor; and 
   recording the vehicle speed to a non-volatile, computer-readable storage medium;   wherein operating the processor to measure the slope comprises operating the processor to:
 normalize the vehicle signature signal in amplitude; 
 identify a first time point, the first time point being an earliest time point at which the amplitude of the normalized vehicle signature signal exceeds a threshold value; and 
 calculating the slope of the normalized vehicle signature signal at the first time point; 
   wherein the threshold value is about 0.4.   
     
     
         13 . The method of  claim 8  further comprising:
 receiving through the hardware interface vehicle classification information; and 
 operating the processor to determine the category in which the vehicle is classified from the vehicle classification information. 
 
     
     
         14 - 20 . (canceled)