Apparatus and method for motion detection and tracking of objects in a region for collision avoidance utilizing a real-time adaptive probabilistic neural network

• US 5,613,039 A
• Filed: 01/03/1994
• Issued: 03/18/1997
• Est. Priority Date: 01/31/1991
• Status: Expired due to Fees

• Current Assignee: AIL Systems, Inc. (ITT, Inc.)
• Original Assignee: AIL Systems, Inc. (ITT, Inc.)
• Inventors: Thompson, James P., Wang, C. David
• Primary Examiner(s): Downs, Robert W.
• Application Number:
  US08/177,359
• Time in Patent Office: 1,170 Days
• Field of Search: 342/28, 342/70, 342/90, 342/95, 342/118, 342/129, 342/195, 342/196, 342/71, 395/22, 395/21, 395/27, 395/11, 395/24, 364/516, 364/517, 340/436
• US Class Current: 706/24
• CPC Class Codes: G01J 3/453 by correlation of the amplitudes
• CPC Class Codes: G01S 13/24 using frequency agility of carrier wave
• CPC Class Codes: G01S 13/348 using square or rectangular modulation, e.g. diplex radar for ranging over short distances
• CPC Class Codes: G01S 13/726 Multiple target tracking
• CPC Class Codes: G01S 13/931 of land vehicles
• CPC Class Codes: G01S 17/95 for meteorological use
• CPC Class Codes: G01S 2013/93271 in the front of the vehicles
• CPC Class Codes: G01S 2013/93272 in the back of the vehicles
• CPC Class Codes: G01S 2013/93274 on the side of the vehicles
• CPC Class Codes: G01S 7/021 Auxiliary means for detecting or identifying radar signals or the like, e.g. radar jamming signals
• CPC Class Codes: G01S 7/2883 using FFT processing
• CPC Class Codes: G01S 7/417 involving the use of neural networks
• CPC Class Codes: G06F 18/2414 Smoothing the distance, e.g. radial basis function networks [RBFN]
• CPC Class Codes: G06N 3/047 Probabilistic or stochastic networks
• CPC Class Codes: G06N 3/049 Temporal neural networks, e.g. delay elements, oscillating neurons or pulsed inputs
• CPC Class Codes: Y02A 90/10 Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation

Abstract

Apparatus for motion detection and tracking of objects in a region for collision avoidance includes a signal transmitter which provides first and second detection signals for at least partial reflection by an object located in a spatial region. The apparatus further includes a signal receiver for receiving the deflected first and second detection signals corresponding to first and second object parameter data signals. The apparatus further includes a Fourier transform circuit for receiving the first and second object parameter data signals and providing first and second Fourier transform object parameter data signals. The apparatus further includes a probabilistic neural network for receiving and sorting the first and second Fourier transform object parameter data signals without the use of a priori training data.

Claims

1. Apparatus for detecting and tracking motion of objects in a region for collision avoidance comprising:

a signal transmitter having an input port and an output port, the signal transmitter being responsive to first and second detection signals at the input port and transmitting the first and second detection signals from the output port to a spatial region, the first and second detection signals at least partially reflecting off at least one object located in the spatial region, the first detection signal having a first signal frequency and the second detection signal having a second signal frequency, the first signal frequency being substantially distinct from the second signal frequency,

a signal receiver being electrically coupled to the transmitter, the receiver having an input port and an output port, the signal receiver being responsive to the at least partially reflected first and second detection signals respectively corresponding to first and second object parameter data signals,

a probabilistic neural network processor being electrically coupled to the signal receiver and being responsive to the first and second object parameter data signals, the probabilistic neural network providing an output signal indicative of the proximity of an object detected in the spatial region;

said probabilistic neural network including;

(a) a plurality of cluster processor circuits being responsive to the first and second object parameter data signals, each cluster processor circuit generating an output signal representing a probability density function estimation value corresponding to the received first and second object parameter data signals, each cluster processor circuit including;

(1) an input buffer memory circuit, the input buffer memory circuit having a plurality of serially connected registers for storing the first and second object parameter data signals assigned to a respective cluster processor circuit;

(2) a plurality of processing elements, each of the processing elements being coupled to a corresponding register of the input buffer memory circuit and being responsive to assigned first and second object parameter data signals stored in the input buffer memory circuit, each of the processing elements further being responsive to current unassigned first and second object parameter data signals, each processing element providing an output signal;

(3) a plurality of exponential function circuits, each of the exponential function circuits being coupled to a corresponding processing element, each exponential function circuit performing an exponential function on the output signal of each processing element and providing an output signal in response thereto; and

(4) a summation circuit coupled to each of the exponential function circuits of the respective cluster processor circuit, the summation circuit being responsive to the output signals from the exponential function circuits, performing an addition function thereon and providing an output signal representing a probability density function estimation value for each unassigned first and second object parameter data signal;

(b) a decision logic circuit, the decision logic circuit being coupled to the summation circuit of each cluster processor circuit, the decision logic circuit comparing the output signal of each summation circuit of the corresponding cluster processor circuit with at least first and second threshold value signals, and providing a decision address signal in response thereto;

(c) a switching circuit, the switching circuit being coupled to the decision logic circuit and to each of the cluster processor circuits and further being responsive to current unassigned first and second object parameter data signals and assigning the current unassigned first and second object parameter data signals to a respective cluster processor circuit for storage in the input buffer memory circuit of the respective cluster processor circuit in response to the decision address signal from the decision logic circuit, the switching circuit being electrically coupled to the input buffer memory circuit of each cluster processor circuit, the switching circuit assigning the current unassigned first and second object parameter data signal to the input buffer memory circuit of a currently operating cluster processor circuit if the summation circuit output signal representing a probability density function estimation value of the currently operating cluster processor circuit is at least equal to the first threshold value signal, the switching circuit assigning the current unassigned first and second object parameter data signal to the input buffer memory circuit of a newly operating cluster processor which was previously a non-operating cluster processor if the summation circuit output signal representing the probability density function estimation value of each currently operating cluster processor circuit is less than or equal to the second threshold value signal; and

(d) a storage register circuit, the storage register circuit being electrically coupled to the switching circuit, the switching circuit providing the current unassigned first and second object parameter data signals to the storage register circuit when the output signal of the summation circuit of each cluster processor circuit is less than the first threshold value signal and greater than the second threshold value signal, the unassigned first and second object parameter data signals being provided to the storage register circuit for later detailed analysis by the probabilistic neural network.

2. Apparatus as defined by claim 1, which further comprises:

a lowpass filter electrically coupled to the signal receiver, the lowpass filter substantially permitting passage of at least a portion of the first and second object parameter data signals less than a first frequency, the lowpass filter being responsive to the first and second object parameter data signals from the receiver and providing a lowpass filter output signal, the lowpass filter output signal being provided to the probabilistic neural network processor.

3. Apparatus as defined by claim 1, which further comprises:

an analog to digital converter, the analog to digital converter being electrically coupled to the signal receiver, the analog to digital converter being responsive to the first and second object parameter data signals and providing a plurality of digital first and second object parameter data signals, the plurality of digital first and second object parameter data signals being provided to the probabilistic neural network processor.

4. Apparatus as defined in claim 1, which further comprises:

a Fourier transform circuit being electrically coupled to the signal receiver, the Fourier transform circuit being responsive to the first and second object parameter data signals, the Fourier transform circuit providing a first and second Fourier transform object parameter data signals representing a spectral waveshape of the first object parameter data signal and the second object parameter data signal.

5. Apparatus as defined by claim 4, which further comprises:

a peak detector electrically coupled to the Fourier transform circuit, the peak detector receiving the first Fourier transform object parameter data signal, the peak detector providing a peak detector output signal indicative of the amplitude of the first object parameter data signal at a peak amplitude of the Fourier transform output parameter data signal, the peak detector output signal being provided to the probabilistic neural network processor, the peak detector output signal being indicative of a velocity of the object detected in the spatial region.

6. Apparatus as defined by claim 5, which further comprises:

a subtractor coupled to the Fourier transform circuit and the peak detector, the subtractor being responsive to the peak detector output signal and the second Fourier transform object parameter detector signal, the subtractor providing a subtractor output signal indicative of the difference between the peak detector output signal and the second Fourier transform output parameter data signal, the subtractor output signal representing a range of the object detected in the spatial region.

7. Apparatus as defined in claim 1, which further comprises:

a signal generator electrically coupled to the input port of the signal transmitter, the signal generator providing at least the first detection signal and the second detection signal to the signal transmitter for transmitting to the spatial region.

8. Apparatus as defined in claim 1, which further comprises:

an antenna electrically coupled to the signal transmitter output port and the signal receiver input port, the antenna being responsive to the first and second detection signals from the signal transmitter, the antenna transmitting the first and second detection signals to the spatial region, the antenna being responsive to the at least partially reflected first and second detection signals.

9. Apparatus as defined in claim 1, which further comprises:

switching means electrically coupled to the signal receiver output port for switching the first and second object parameter data signals, the switching means having an input port and at least first and second output ports, the switching means input port receiving both the first and second object parameter data signals, the switching means first output port providing the first object parameter data signal and the switching means second output port providing the second object parameter data signal.

10. Apparatus for detecting and tracking motion as defined by claim 1, wherein the first and second detection signals are continuous wave signals.

11. An apparatus as defined in claim 1, wherein said first threshold value signal is about 70% and said second threshold value signal is about 10%.

12. Apparatus for detecting and tracking motion of objects in a region for collision avoidance comprising:

a signal transmitter having an input port and an output port, the signal transmitter being responsive to first and second detection signals at the input port and transmitting the first and second detection signals from the output port to a spatial region, the first and second detection signals at least partially reflecting off at least one object located in the spatial region, the first detection signal having a first signal frequency and the second detection signal having a second signal frequency, the first signal frequency being substantially distinct from the second signal frequency,

a signal receiver being electrically coupled to the transmitter, the receiver having an input port and an output port, the signal receiver being responsive to the at least partially reflected first and second detection signals respectively corresponding to first and second object parameter data signals,

a Fourier transform circuit being electrically coupled to the signal receiver and being responsive to the first and second object parameter data signals, the Fourier transform circuit providing first and second Fourier transform object parameter data signals representing a spectral waveshape of the first object parameter data signal and the second object parameter data signal, and

a probabilistic neural network processor being electrically coupled to the Fourier transform circuit and being responsive to the first and second object parameter data signals, the probabilistic neural network providing an output signal indicative of the proximity of an object detected in the spatial region, the probabilistic neural network comprising;

(a) a plurality of cluster processor circuits being responsive to the first and second Fourier transform object parameter data signals, each cluster processor circuit generating an output signal representing a probability density function estimation value corresponding to the received first and second Fourier transform object parameter data signals, each cluster processor circuit including;

(1) an input buffer memory circuit, the input buffer memory circuit having a plurality of serially connected registers for storing the first and second Fourier transform object parameter data signals assigned to a respective cluster processor circuit;

(2) a plurality of processing elements, each of the processing elements being coupled to a corresponding register of the input buffer memory circuit and being responsive to assigned first and second Fourier transform object parameter data signals stored in the input buffer memory circuit, each of the processing elements further receiving current unassigned first and second Fourier transform object parameter data signals, each processing element providing an output signal;

(3) a plurality of exponential function circuits, each of the exponential function circuits being coupled to a corresponding processing element, each exponential function circuit performing an exponential function on the output signal of each processing element and providing an output signal in response thereto; and

(4) a summation circuit coupled to each of the exponential function circuits of the respective cluster processor circuit, the summation circuit being responsive to the output signals from the exponential function circuits, performing an addition function thereon and providing an output signal representing a probability density function estimation value for each unassigned first and second Fourier transform object parameter data signal;

(b) a decision logic circuit, the decision logic circuit being coupled to the summation circuit of each cluster processor circuit, the decision logic circuit comparing the output signal of each summation circuit of the corresponding cluster processor circuit with at least first and second threshold value signals, and providing a decision address signal in response thereto;

(c) a switching circuit, the switching circuit being coupled to the decision logic circuit and to each of the cluster processor circuits and further being responsive to current unassigned first and second Fourier transform object parameter data signals and assigning the current unassigned first and second Fourier transform object parameter data signals to a respective cluster processor circuit for storage in the input buffer memory circuit of the respective cluster processor circuit in response to the decision address signal from the decision logic circuit, the switching circuit being electrically coupled to the input buffer memory circuit of each cluster processor circuit, the switching circuit assigning the current unassigned first and second Fourier transform object parameter data signals to the input buffer memory circuit of a currently operating cluster processor circuit if the summation circuit output signal representing a probability density function estimation value of the currently operating cluster processor circuit is at least equal to the first threshold value signal, the switching circuit assigning the current unassigned first and second Fourier transform object parameter data signals to the input buffer memory circuit of a newly operating cluster processor which was previously a non-operating cluster processor if the summation circuit output signal representing the probability density function estimation value of each currently operating cluster processor circuit is less than or equal to the second threshold value signal;

anal

(d) a storage register circuit, the storage register circuit being electrically coupled to the switching circuit, the switching circuit providing the current unassigned first and second Fourier transform object parameter data signals to the storage register circuit when the output signal of the summation circuit of each cluster processor circuit is less than the first threshold value signal and greater than the second threshold value signal, the unassigned first and second Fourier transform object parameter data signals being provided to the storage register circuit for later detailed analysis by the probabilistic neural network.

13. Apparatus for detecting and tracking motion as defined by claim 12, wherein the first and second detection signals are continuous wave signals.

14. An apparatus as defined in claim 12, wherein said first threshold value signal is about 70% and said second threshold value signal is about 10%.

15. A method of detecting and tracking motion of objects in a region for collision avoidance which comprises:

a) transmitting first and second detection signals having distinct frequencies for at least partial reflection by at least one object in a spatial region, the first detection signal having a first signal frequency and the second detection signal having a second signal frequency, the first signal frequency being substantially distinct from the second signal frequency;

b) detecting the first and second detection signals at least partially reflected by the object, the first and second detection signals corresponding to first and second object parameter data signals;

c) providing the first and second object parameter data signals to a probabilistic neural network for sorting the first and second object parameter data signals without the use of a priori training data, step (c) including the additional sub-steps of;

i) inputting current unassigned first and second object parameter data signals into at least one of a plurality of processing elements contained within a plurality of currently operating cluster processor circuits;

ii) generating a first signal representing a probability density function estimation value in response to the first and second object parameter data signal of each currently operating cluster processor circuit of the plurality of cluster processors, the probability density function estimation value signal being generated using the current unassigned first and second object parameter data signals and using a plurality of assigned first and second object parameter data signals stored in an input buffer memory circuit of each currently operating cluster processor circuit;

iii) comparing the first signal representing the probability density function estimation value generated by each currently operating cluster processor circuit to at least first and second threshold value signals in a decision logic circuit;

iv) generating a decision address signal in response to the comparison of the first signal and the first and second threshold value signals, the decision address signal being provided to a switching circuit, the decision address signal denoting a currently operating cluster processor circuit when the first signal representing the probability density function estimation value is at least equal to the first threshold value signal, the decision address signal denoting and activating a non-operating cluster processor circuit when the first signal representing the probability density function estimation value is at most equal to the second threshold value signal;

v) assigning the current unassigned first and second object parameter data signal from the switching circuit to the cluster processor circuit corresponding to the decision address signal provided to the switching circuit;

vi) storing the current unassigned first and second object parameter data signal in the input buffer memory circuit of the cluster processor circuit according to the decision address signal received by the switching circuit; and

vii) denoting a storage register circuit and assigning the current unassigned first and second object parameter data signal from the switching circuit to the storage register circuit for temporary storage therein when the first signal, representing the probability density function estimation value, is less than the first threshold value signal and greater than the second threshold value signal, the unassigned first and second object parameter data signal being provided to the storage register circuit for later detailed analysis by the probabilistic neural network.

16. A method as defined by claim 15, which includes the further step of manually setting the level of at least one of the first and second threshold value signals.

17. A method as defined by claim 15, wherein the first threshold value signal is greater in magnitude than the second threshold value signal.

18. A method as defined by claim 15, wherein prior to step (c), the method further includes Fourier transforming the first and second object parameter data signals to provide Fourier transformed first and second object parameter data signals.

19. A method of detecting and tracking motion as defined by claim 15, wherein the first and second detection signals are continuous wave signals.

20. A method as defined in claim 15, wherein in step (c)(iii), said first threshold value signal is about 70% and said second threshold value signal is about 10%.