System and method for distinguishing and characterizing motor vehicles for control of automatic drivers

Claim 1

A method for determining characteristics of a motor vehicle comprising the steps of: monitoring a torque signal produced by operation of an engine of a motor vehicle; filtering said torque signal to generate component frequency signals representative of component frequencies within said torque signal, by generating a frequency membership vector signal representative of the component frequencies within said torque signal; and generating characteristic signals representative of relative temperature, inertia and horsepower of said motor vehicle in response to determined component frequencies within said torque signal by passing said frequency membership vector signal through a neural network.

Claim 2

A method for determining characteristics of a motor vehicle comprising the steps of: monitoring a torque signal produced by operation of an engine of a motor vehicle; filtering said torque signal to generate component frequency signals representative of component frequencies within said torque signal, by generating a frequency membership vector signal representative of the component frequencies within said torque signal; and generating characteristic signals representative of relative temperature, inertia and horsepower of said motor vehicle in response to determined component frequencies within said torque signal by passing said frequency membership vector signal through a neural network.

Claim 10

A method for determining characteristics of and controlling a motor vehicle comprising the steps of: monitoring a torque signal produced by operation of an engine of a motor vehicle; filtering said torque signal to generate component frequency signals representative of component frequencies within said torque signal; generating characteristic signals representative of characteristics of said motor vehicle in response to determined component frequencies within said torque signal; utilizing said characteristic signals to control a robot vehicle driver and thereby operation of said motor vehicle while said motor vehicle is on a dynamometer, robot vehicle driver control including generation of an anticipated throttle lag signal generated in response to said characteristics signals; generating speed schedule signals representative of vehicle speed over time for said motor vehicle; generating brake and throttle action signals which are passed to said robot vehicle driver in response to a requested throttle position signal, a requested brake position signal and said anticipated throttle lag signal; combining said speed schedule signals with a dynamometer roll speed signal, said characteristic signals, and said brake and throttle action signals to generate said requested throttle position signal and said requested brake position signal.

Claim 11

A method for determining characteristics of and controlling a motor vehicle comprising the steps of: monitoring a torque signal produced by operation of an engine of a motor vehicle; filtering said torque signal to generate component frequency signals representative of component frequencies within said torque signal; generating characteristic signals representative of characteristics of said motor vehicle in response to determined component frequencies within said torque signal; utilizing said characteristic signals to control a robot vehicle driver and thereby operation of said motor vehicle while said motor vehicle is on a dynamometer, robot vehicle driver control including generation of an anticipated throttle lag signal generated in response to said characteristics signals; generating speed schedule signals representative of vehicle speed over time for said motor vehicle; generating brake and throttle action signals which are passed to said robot vehicle driver in response to a requested throttle position signal, a requested brake position signal and said anticipated throttle lag signal; combining said speed schedule signals with a dynamometer roll speed signal, said characteristic signals, and said brake and throttle action signals to generate said requested throttle position signal and said requested brake position signal.

Claim 17

A system for characterizing a motor vehicle and controlling said motor vehicle via a robot vehicle driver, said motor vehicle being on a dynamometer and said system comprising: a fuzzy spectral filter for filtering a torque signal produced by operation of an engine of a motor vehicle, said fuzzy spectral filter producing component frequencies within said torque signal; a neural network for generating characteristic signals representative of characteristics of said motor vehicle in response to determined component frequencies within said torque signal; a neural network for generating an anticipated throttle lag signal in response to said characteristic signals; a source of speed schedule signals representative of speed of said motor vehicle over time; a fuzzy scheduler for generating brake and throttle action signals which are passed to a robot vehicle driver in response to a requested throttle signal, a requested brake position signal and said anticipated throttle lag signal; and a neural network for combining said speed schedule signals with a dynamometer roll speed signal, said characteristic signals, and said brake and throttle action signals to generate said requested throttle position signal and said requested brake position signal.

Claim 18

A system for characterizing a motor vehicle and controlling said motor vehicle via a robot vehicle driver, said motor vehicle being on a dynamometer and said system comprising: a fuzzy spectral filter for filtering a torque signal produced by operation of an engine of a motor vehicle, said fuzzy spectral filter producing component frequencies within said torque signal; a neural network for generating characteristic signals representative of characteristics of said motor vehicle in response to determined component frequencies within said torque signal; a neural network for generating an anticipated throttle lag signal in response to said characteristic signals; a source of speed schedule signals representative of speed of said motor vehicle over time; a fuzzy scheduler for generating brake and throttle action signals which are passed to a robot vehicle driver in response to a requested throttle signal, a requested brake position signal and said anticipated throttle lag signal; and a neural network for combining said speed schedule signals with a dynamometer roll speed signal, said characteristic signals, and said brake and throttle action signals to generate said requested throttle position signal and said requested brake position signal.

Claim 21

A method for determining characteristics of a motor vehicle on a dynamometer comprising the steps of: monitoring a dynamometer torque signal produced by operation of an engine of a motor vehicle; filtering said dynamometer torque signal to generate component frequency signals representative of component frequencies within said torque signal by generating a frequency membership vector signal representative of the component frequencies within said torque signal; and generating relative temperature, inertia and horsepower signals representative of characteristics of said motor vehicle in response to determined component frequencies within said dynamometer torque signal by passing said frequency membership vector signal through a neural network.

Claim 22

A method for determining characteristics of a motor vehicle on a dynamometer comprising the steps of: monitoring a dynamometer torque signal produced by operation of an engine of a motor vehicle; filtering said dynamometer torque signal to generate component frequency signals representative of component frequencies within said torque signal by generating a frequency membership vector signal representative of the component frequencies within said torque signal; and generating relative temperature, inertia and horsepower signals representative of characteristics of said motor vehicle in response to determined component frequencies within said dynamometer torque signal by passing said frequency membership vector signal through a neural network.