Predictive collision sensing system

Claim 1

A predictive collision sensing system for a vehicle, comprising: a. a radar transmitter/receiver module for illuminating a zone proximate the vehicle with a beam of electromagnetic energy and for receiving portions of said beam of electromagnetic energy reflected by one or more objects illuminated by said electromagnetic energy thereby providing an estimate of the distance to said target relative to the vehicle; b. a signal processor operatively coupled to said radar transmitter/receiver module; c. a means under control of said signal processor for scanning said beam of electromagnetic energy over an azimuth range in accordance with an azimuth coverage pattern, wherein said azimuth coverage pattern is continuous; d. a target tracking algorithm implemented by said signal processor for tracking said one or more objects relative to the vehicle; e. a clustering algorithm implemented by said signal processor for grouping separate tracks of said one or more objects; f. a target size estimating algorithm implemented by said signal processor for estimating the azimuthal extent of said one or more objects tracked by said target tracking algorithm; and g. a threat assessment algorithm implemented by said signal processor for assessing the threat of said one or more objects to the vehicle, whereby said signal processor controls one or more vehicular devices responsive to said assessment of threat, so as to enhance the safety of the occupant of the vehicle.

Claim 19

A method of predicting a crash for a vehicle and for controlling one or more vehicular devices responsive thereto, comprising: a. measuring the range of one or more objects relative to the vehicle using an azimuthally scanned radar beam; b. calculating the trajectory of said one or more objects from said range measurements; c. clustering separate tracks of said one or more objects; d. measuring the azimuthal extent of said one or more objects from said range measurements and from the azimuthal scan angle of said radar beam; e. assessing the threat of said one or more objects to the vehicle from a combination of said trajectory and said azimuthal extent of said one or more objects; and f. controlling one or more vehicular devices responsive to said assessment of threat, so as to enhance the safety of the occupant of the vehicle.

Claim 20

A method of predicting a crash for a vehicle and for controlling one or more vehicular devices responsive thereto, comprising: a. measuring the range of one or more objects relative to the vehicle using an azimuthally scanned radar beam; b. calculating the trajectory of said one or more objects from said range measurements; c. clustering separate tracks of said one or more objects; d. measuring the azimuthal extent of said one or more objects from said range measurements and from the azimuthal scan angle of said radar beam; e. assessing the threat of said one or more objects to the vehicle from a combination of said trajectory and said azimuthal extent of said one or more objects; and f. controlling one or more vehicular devices responsive to said assessment of threat, so as to enhance the safety of the occupant of the vehicle.

Claim 32

A method of predicting a crash for a vehicle and for controlling one or more vehicular devices responsive thereto, comprising: a. measuring the range of one or more objects relative to the vehicle using an azimuthally scanned radar beam; b. measuring the velocity of one or more objects relative to the vehicle using said azimuthally scanned radar beam, wherein the operations of measuring the range and velocity of one or more objects relative to the vehicle using said azimuthally scanned radar beam comprises: i. generating a continuous wave RF signal comprising a repetitive sequence of uniformly spaced frequencies so as to form a transmitted signal; ii. illuminating the object with said continuous wave RF signal; iii. receiving the component of said RF signal reflected by said one or more objects so as to form a received signal; iv. mixing said received signal with said transmitted signal so as to form a complex amplitude frequency difference signal from the difference in frequency therebetween; v. comparing the real part of said complex frequency difference signal with a plurality of monotonically increasing threshold values, wherein adjacent threshold values corresponds to the bounds of an associated range cell, vi. storing the complex amplitudes of said received signal for each said associated range cell so as to form an associated list of complex amplitudes; vii. calculating the Doppler shift for each said associated range cell from a spectral analysis of said associated list of complex amplitudes; and viii. collecting said Doppler shift calculations for all said associated range cells so as to form a range-Doppler map for said one or more objects, thereby providing the distance and velocity of said one or more objects relative to the vehicle; c. calculating the trajectory of said one or more objects from said range measurements; d. clustering separate tracks of said one or more objects; e. measuring the azimuthal extent of said one or more objects from said range measurements and from the azimuthal scan angle of said radar beam; f. assessing the threat of said one or more objects to the vehicle from a combination of said trajectory and said azimuthal extent of said one or more objects; and g. controlling one or more vehicular devices responsive to said assessment of threat, so as to enhance the safety of the occupant of the vehicle.

Claim 33

A method of predicting a crash for a vehicle and for controlling one or more vehicular devices responsive thereto, comprising: a. measuring the range of one or more objects relative to the vehicle using an azimuthally scanned radar beam; b. measuring the velocity of one or more objects relative to the vehicle using said azimuthally scanned radar beam, wherein the operations of measuring the range and velocity of one or more objects relative to the vehicle using said azimuthally scanned radar beam comprises: i. generating a continuous wave RF signal comprising a repetitive sequence of uniformly spaced frequencies so as to form a transmitted signal; ii. illuminating the object with said continuous wave RF signal; iii. receiving the component of said RF signal reflected by said one or more objects so as to form a received signal; iv. mixing said received signal with said transmitted signal so as to form a complex amplitude frequency difference signal from the difference in frequency therebetween; v. comparing the real part of said complex frequency difference signal with a plurality of monotonically increasing threshold values, wherein adjacent threshold values corresponds to the bounds of an associated range cell, vi. storing the complex amplitudes of said received signal for each said associated range cell so as to form an associated list of complex amplitudes; vii. calculating the Doppler shift for each said associated range cell from a spectral analysis of said associated list of complex amplitudes; and viii. collecting said Doppler shift calculations for all said associated range cells so as to form a range-Doppler map for said one or more objects, thereby providing the distance and velocity of said one or more objects relative to the vehicle; c. calculating the trajectory of said one or more objects from said range measurements; d. clustering separate tracks of said one or more objects; e. measuring the azimuthal extent of said one or more objects from said range measurements and from the azimuthal scan angle of said radar beam; f. assessing the threat of said one or more objects to the vehicle from a combination of said trajectory and said azimuthal extent of said one or more objects; and g. controlling one or more vehicular devices responsive to said assessment of threat, so as to enhance the safety of the occupant of the vehicle.