Target route predicting apparatus utilizing characteristic parameters of terrain information

• US 5,408,414 A
• Filed: 03/04/1992
• Issued: 04/18/1995
• Est. Priority Date: 03/08/1991
• Status: Expired due to Fees

• Current Assignee: Mitsubishi Denki Kabushiki Kaisha (Mitsubishi Electric Corporation)
• Original Assignee: Mitsubishi Denki Kabushiki Kaisha (Mitsubishi Electric Corporation)
• Inventors: Nomoto, Kohei, Oh-Hashi, Yoshimasa
• Primary Examiner(s): Chin, Gary
• Application Number:
  US07/846,520
• Time in Patent Office: 1,140 Days
• Field of Search: 364/443, 364/444, 364/449, 364/460, 364/461, 340/436, 340/988, 340/990, 340/995, 73/178 R, 342/450, 342/451, 342/455, 342/457
• US Class Current: 701/209
• CPC Class Codes: G01C 21/005 with correlation of navigation data from several sources, e.g. map or contour matching G01C21/30 takes precedence
• CPC Class Codes: G05D 1/106 Change initiated in response to external conditions, e.g. avoidance of elevated terrain or of no-fly zones
• CPC Class Codes: G08G 5/0086 for monitoring terrain radar or analogous systems specially adapted for terrain avoidance G01S13/935
• CPC Class Codes: G08G 5/045 Navigation or guidance aids, e.g. determination of anti-collision manoeuvers radar or analogous systems specially adapted for anti-collision between aircraft G01S13/933

Abstract

A route predicting apparatus for monitoring a flying target and predicting a route thereof. Even if the target is hidden by mountains or other obstacles thus preventing observation for a certain time, a reliable route prediction is carried out by utilizing the fact that a route of the target is restricted by geographical features. The apparatus comprises an observation unit for observing a target and outputting an observed value thereof, a terrain information unit for outputting characteristic parameters of stored terrain information, a prediction unit coupled to receive the observed value and the characteristic parameters of terrain information for performing a fuzzy inference to output a predicted observed value of the target, and a learning adjustment unit coupled to receive an error signal between a predicted observed value at the current time predicted a unit time before and a real observed value at the current time for adjusting the prediction unit by using a neural network.

Claims

1. An apparatus for predicting a predicted route of a target, comprising:

an observing unit for observing the target, having means for determining if the target is observable and means for providing an observed route value of an actual route of said target;

terrain information means for providing characteristic parameters of prestored terrain information; and

a predicting unit, coupled to receive said observed route value and said characteristic parameters, having means for inferring the predicted route according to a first and a second mode,

in which the means for inferring operates according to the first mode, when the means for determining determines that the target is observable, and infers the predicted route on the basis of the observed route value and the characteristic parameters and

in which the means for inferring operates according to the second mode, when the means for determining determines that the target is unobservable, and infers the predicted route on the basis of a prior predicted route and the characteristic parameters.

2. The apparatus according to claim 1, wherein said means for inferring includes means for adjusting the inference in response to an adjusting signal and wherein said predicting unit further comprises:

delay means, receiving the predicted route, for delaying the predicted route by a unit time to output a delayed predicted route

subtracting means, coupled to receive the delayed predicted route and the observed route value, for providing an error signal that is indicative of an accuracy of the predicted route in comparison to the actual route; and

neural network means, receiving the error signal, for providing the adjusting signal so that the means for inferring gradually improves the accuracy.

3. The apparatus according to claim 2, wherein said terrain information means includes a memory for prestoring said terrain information and an extracting means for extracting the characteristic parameters of the terrain information from said memory.

4. An apparatus for predicting a predicted state of a target for a unit time later than the current time, comprising:

an observing unit having a sensor for observing the target and having a processor processing signals from the sensor to provide an observed value indicative of an actual route of the target;

memory means for prestoring terrain information;

calculating means, responsive to the terrain information and an estimated state of the target for the current time, for calculating a steering amount of the target, the steering amount being indicative of the amount of steering the target is forced to generate in the estimated state because of a restriction of terrain as indicated by the terrain information; and

predicting means, responsive to the observed value and the steering amount, for providing the estimated state of said target for the current time and the predicted state of said target at said unit time later than the current time.

5. The apparatus as claimed in claim 4 wherein said predicting means comprises:

means, responsive to the predicted state of said target, for outputting a predicted observed value;

means, responsive to the observed value and the predicted observed value, for outputting the estimated state of the target for the current time; and

means, responsive to the estimated state of the target and the steering amount, for outputting the predicted state.

6. An apparatus for predicting a predicted state of a target at a unit time later than the current time, comprising:

an observing unit having a sensor for observing the target and having a processor processing signals from the sensor to provide an observed value indicative of an actual route of the target;

memory means for prestoring terrain information;

calculating means, responsive to the terrain information from said memory means and a predicted state of said target for the current time, for calculating a steering amount of said target, the steering amount being indicative of the amount of steering the target is forced to generate in the predicted state for the current time because of a restriction of terrain as indicated by the terrain information; and

predicting means, responsive to the steering amount, the observed value and the predicted state for the current time, for outputting the predicted state of said target for said unit time later than the current time.

7. An apparatus for predicting a predicted state of a target at a unit time later than the current time, comprising:

an observing unit having a sensor for observing the target and having a processor processing signals from the sensor to provide an observed value indicative of an actual route of the target;

memory means for prestoring terrain information;

first predicting means, responsive to the terrain information from said memory means and an estimated current state of said target, for outputting a maneuver signal representative of a predicted amount of maneuver necessary by said target for the purpose of avoiding collision to an obstacle as indicated by the terrain information; and

second predicting means, responsive to the observed value from said observing means and the maneuver signal from said first predicting means, for providing the estimated state of said target and the predicted state of said target at said unit time later than the current time.

8. The apparatus as claimed in claim 7 wherein said second predicting means comprises:

means, responsive to the predicted state of said target, for outputting a predicted observed value;

means, responsive to the observed value and the predicted observed value, for outputting the estimated state of said target; and

means, responsive to the estimated state of said target, for outputting the predicted state at said unit time later than the current time.

9. An apparatus for predicting a route of a target for a unit time later than the current time, comprising:

an observing unit having a sensor for observing the target and having a processor for processing signals from the sensor to provide an observed value indicative of an actual route of the target;

a terrain information unit having a memory with terrain information stored therein and a means for extracting characteristic parameters from the terrain information stored in the memory; and

a predicting unit having a means for inferring and a means for adjusting, the means for inferring, including

a fuzzy inferring element, receiving the characteristic parameters, an adjusting signal from the means for adjusting, the observed value, and a predicted route value for the current time, the fuzzy inferring element operating according to one of a first and a second mode in response to whether the observing unit is able to observe the target,

in which in the first mode, the fuzzy inferring element infers the predicted route value of the target for said unit time later than the current time on the basis of the observed value, the adjusting signal, and the characteristic parameters, and in which in the second mode, the fuzzy inferring element infers the predicted route value of the target for said unit time later than the current time on the basis of the characteristic parameters and the predicted route value of the target for the current time,

wherein the first mode corresponds to the target being observable and the second mode corresponds to the target being unobservable, by the observing unit and

a delay element for storing for one time unit the predicted route value of the target for said unit time later than the current time and for providing a delayed predicted route value, and

the means for adjusting, including

a subtraction element, receiving the delayed predicted route value and the observed value, for providing an error signal indicative of the accuracy of the inference performed by the means for inferring at said unit time prior to the current time, and

a neural network receiving the error signal and producing therefrom the adjusting signal so as to gradually improve the accuracy over time.