Control device for an automobile

• US 5,361,213 A
• Filed: 03/24/1993
• Issued: 11/01/1994
• Est. Priority Date: 02/09/1990
• Status: Expired due to Fees

• Current Assignee: Hitachi, Ltd.
• Original Assignee: Hitachi, Ltd.
• Inventors: Fujieda, Mamoru, Ooyama, Yoshishige, Ohsuga, Minoru, Nogi, Toshiharu
• Primary Examiner(s): Trans, Vincent N.
• Application Number:
  US08/037,732
• Time in Patent Office: 587 Days
• Field of Search: 364/431.08, 364/431.07, 364/431.04, 123/419, 123/436, 395/21, 395/22, 395/905
• US Class Current: 701/111
• CPC Class Codes: B60K 31/0008 including means for detecting potential obstacles in vehicle path
• CPC Class Codes: B60W 2030/206 related or induced by the engine
• CPC Class Codes: B60W 2050/0002 Automatic control, details of type of controller or control system architecture
• CPC Class Codes: B60W 2050/0043 Signal treatments, identification of variables or parameters, parameter estimation or state estimation
• CPC Class Codes: B60W 2050/0057 Frequency analysis, spectral techniques or transforms
• CPC Class Codes: B60W 30/20 Reducing vibrations in the driveline
• CPC Class Codes: B60W 50/00 Details of control systems for road vehicle drive control not related to the control of a particular sub-unit , e.g. process diagnostic or vehicle driver interfaces
• CPC Class Codes: F02B 2075/027 four
• CPC Class Codes: F02D 2041/1433 using a model or simulation of the system
• CPC Class Codes: F02D 41/1405 Neural network control
• CPC Class Codes: F02P 5/045 combined with electronic control of other engine functions, e.g. fuel injection in general F02D37/02
• CPC Class Codes: F16H 2061/0081 Fuzzy logic
• CPC Class Codes: F16H 2061/0084 Neural networks
• CPC Class Codes: F16H 2061/0087 Adaptive control, e.g. the control parameters adapted by learning
• CPC Class Codes: G01L 23/225 circuit arrangements therefor
• CPC Class Codes: G01S 17/931 of land vehicles
• CPC Class Codes: Y10S 706/905 Vehicle or aerospace

Abstract

A control device for an automobile has at least one sensor for detecting an operational state of the automobile. An output from the sensor is applied to a sample and hold circuit which is adapted to provide a time displaced output to a hierarchical group of neural elements. The neural elements operate upon said time displaced signals by varying the weighting of the signals and/or threshold values of the neural elements in accordance with a desired output parameter. The output of the neural computer is applied to a control circuit which is adapted to control an automobile actuator. The hierarchical group of neural elements are provided with numerous forms of different learning functions so that the desired output parameter is achieved. In one such embodiment desired parameters are applied to said neural computer and to a comparator, signals derived from sensors associated with automobile actuators are also applied to said comparator whereby difference signals are outputted from said comparator to control the weighting and/or threshold levels of the neural elements. The invention may be applied to different functions such as controlling air/fuel ratio, ignition timing, knock control, sensing, suspension adjustment, and road surface detection.

Claims

1. A control device for an automobile having an engine knock sensor for detecting engine knock of said automobile engine, and an actuator for controlling said engine knock, said control device including transformation means for transforming signals from said knock sensor into a spatial multi-variable signal consisting of plural bits and to apply said spatial multi-variable signal to a hierarchical group of neural elements, each element providing an output O₁ where ##EQU12## where O_j are inputs, W_ij are weighting factors and θ

is a threshold,

and further including modifying means for adaptively controlling the internal parameters of the neural elements by comparison with actual knock data, and a control means for receiving an output signal from said group of neural elements for controlling said actuator.

2. A control device as claimed in claim 1 wherein said modifying means includes means for varying weighting values of each neural element in the group of neural elements to provide said desired parameter.

3. A control device as claimed in claim 1 wherein the modifying means includes a threshold modifying means for modifying a threshold value of said neural element in said hierarchical group of neural elements to provide said desired parameter.

4. A control device as claimed in claim 1 wherein said group of neural elements comprises an input layer of neural elements, an intermediate layer of neural elements and an output layer of neural elements, and difference means are connected to each neural element of said output layer of neural elements, said difference means also being connected to receive an input signal to an associated neural element of said input layer of neural elements.

5. A control device as claimed in claim 1 wherein a controller is provided for providing predetermined, desired, sensor values to said hierarchical group of neural elements and an error feedback means is provided for receiving signals from said controller and from said at least one sensor, said error feedback means modifying the output of the neural elements in said hierarchical group of neural elements whereby the difference in signals applied to said error feedback means are brought to zero.

6. A control device as claimed in claim 5 wherein the at least one sensor comprises a light emitter and a light detector array for supplying signals to said hierarchical group of neural elements.

7. A control device as claimed in claim 1 wherein a plurality of sensors are provided and a plurality of control means are provided, and driver actuable means are provided for controlling the output of the hierarchical group of neural elements whereby operational parameters of the automobile may be varied in dependence upon said drivers preference.

8. A control device for an automobile having an engine knock sensor for detecting engine knock of said automobile engine, and an actuator for controlling said engine knock, said control device including

(a) transformation means for transforming time series signals from said knock sensor into a spatial multi-variable signal consisting of plural bits, said transformation means having a sampling and holding means which starts sampling and holding on the basis of a sample start signal which is related to an instant of ignition and carries out the sampling and holding process for a predetermined period of time, whereby n old samples are replaced by n new samples at each predetermined period of time, and to apply said spatial multi-variable signal to

(b) an input layer of n neural elements of a hierarchical group of neural elements, each element providing an output O₁ where ##EQU13## where O_j are inputs, W_ij are weighting factors and θ

is a threshold,

(c) means for measuring actual knock data and for producing an output signal representative thereof,

(d) modifying means for adaptively controlling the internal parameters of the neural elements by comparison with the actual knock data measuring means output signal, and

(e) control means for receiving an output signal from said group of neural elements for controlling said actuator.

9. A control device as claimed in claim 8 wherein said modifying means includes means for varying weighting values of each neural element in the group of neural elements to provide said desired parameter.

10. A control device as claimed in claim 8 wherein the modifying means includes a threshold modifying means for modifying a threshold value of said neural element in said hierarchical group of neural elements to provide said desired parameter.

11. A control device as claimed in claim 8 wherein said group of neural elements comprises said input layer of neural elements, an intermediate layer of neural elements and an output layer of neural elements, and difference means are connected to each neural element of said output layer of neural elements, said difference means also being connected to receive an input signal to an associated neural element of said input layer of neural elements.

12. A control device as claimed in claim 8 wherein a controller is provided for providing predetermined, desired, sensor values to said hierarchical group of neural elements and an error feedback means is provided for receiving signals from said controller and from said at least one sensor, said error feedback means modifying the output of the neural elements in said hierarchical group of neural elements whereby the difference in signals applied to said error feedback means are brought to zero.

13. A control device as claimed in claim 12 wherein the at least one sensor comprises a light emitter and a light detector array for supplying signals to said hierarchical group of neural elements.

14. A control device as claimed in claim 8 wherein a plurality of sensors are provided and a plurality of control means are provided, and driver actuable means are provided for controlling the output of the hierarchical group of neural elements whereby operational parameters of the automobile may be varied in dependence upon said drivers preference.

15. A control device for an automobile having at least one sensor adapted to detect an operational state of said automobile, and an actuator for controlling said operational state, said control device including:

(a) a transformation means for transforming signals in time series from said at least one sensor into a spatial multi-variable signal, said transformation means having a sampling and holding means which starts sampling and holding on the basis of a sample hold start signal relating to an instant of ignition as its starting point, and carries out the sampling and holding process for a predetermined period of time, the spatial multi-variable signal consisting of plural bits (12a, 12b, 12c, - - - 12n) in sequence of input order and being renewed at each predetermined period of time so that n old samples are replaced by n new samples at each predetermined period of time;

(b) a hierarchical group of neural elements having an input layer of n neural elements for receiving said spatial multi-variable signal;

(c) modifying means for adaptively controlling the neural elements in accordance with a desired parameter; and

(d) a control means for receiving an output signal from said group of neural elements for controlling said actuator.

16. A control device as claimed in claim 15 wherein said modifying means includes means for varying weighting values of each neural element in the group of neural elements to provide said desired parameter.

17. A control device as claimed in claim 15 wherein the modifying means includes a threshold modifying means for modifying a threshold value of said neural element in said hierarchical group of neural elements to provide said desired parameter.

18. A control device as claimed in claim 15 wherein a plurality of sensors are provided for detecting an operational state of respective components in an automobile and at least one actuator is provided for controlling each said component of said automobile.

19. A control device as claimed in claim 18 wherein there is provided a knock sensor connected to apply signals therefrom to said transformation means, said group of neural elements comprising;

(e) said input layer of neural elements;

(f) an intermediate layer of neural elements;

(g) an output of neural elements;

(h) difference means connected to each neural element of said output layer of neural elements; and

(i) said difference means also being connected to receive an input signal to an associated neural element of said input layer of neural elements.

20. A control device as claimed in claim 15 wherein:

a controller is provided for providing predetermined, desired, sensor values to said hierarchical group of neural elements; and

an error feedback means is provided for receiving signals from said controller and from said at least one sensor, said error feedback means being adapted to modify the output of the neural elements in said hierarchical group of neural elements whereby the difference in signals applied to said error feedback means are brought to zero.

21. A control device as claimed in claim 20 wherein the at least one sensor comprises a light emitter and a light detector array adapted to supply signals to said hierarchical group of neural elements.

22. A control device as claimed in claim 15 wherein a plurality of sensors are provided and a plurality of control means are provided, and driver actuable means are provided adapted to control the output of the hierarchical group of neural elements whereby operational parameters of the automobile may be varied in dependence upon said drivers preference.