Automatic train operation apparatus incorporating security function with improved reliability

• US 5,583,769 A
• Filed: 03/16/1995
• Issued: 12/10/1996
• Est. Priority Date: 09/21/1990
• Status: Expired due to Term

• Current Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Original Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Inventors: Saitoh, Hiroo
• Primary Examiner(s): Chin, Gary
• Assistant Examiner(s): NOT, DEFINED
• Application Number:
  US08/406,246
• Time in Patent Office: 635 Days
• Field of Search: 364/426.01, 364/424.02, 364/184, 364/436, 364/513, 244/182 B, 244/34 CT, 371/9, 375/107, 340/825.17, 395/11
• US Class Current: 701/79
• CPC Class Codes: B61L 15/0062 On-board target speed calculation or supervision
• CPC Class Codes: G05B 9/03 with multiple-channel loop, i.e. redundant control systems
• CPC Class Codes: G06F 11/0796 Safety measures, i.e. ensuring safe condition in the event of error, e.g. for controlling element
• CPC Class Codes: G06F 11/181 Eliminating the failing redundant component
• CPC Class Codes: G06F 11/182 based on mutual exchange of the output between redundant processing components
• CPC Class Codes: G06F 11/184 where the redundant components implement processing functionality
• CPC Class Codes: G06F 11/187 Voting techniques
• CPC Class Codes: G06N 3/045 Combinations of networks

Abstract

An automatic train operation apparatus capable of realizing an optimal train operation with an improved reliability. The apparatus includes: an ATO/C system including a plurality of ATO/C units redundantly provided, each ATO/C unit having a fail safe configuration formed by a plurality of execution processors and a supervisor processor for monitoring normal operations of the execution processors, each execution processor being capable of executing an automatic train operation program, and all of the execution processors in the plurality of ATO/C units executing an identical automatic train operation program simultaneously; and a majority logic circuit for selecting an output obtained by a majority of the execution processors in the ATO/C units of the ATO/C system as a control command output for controlling a train operation. Each execution processor in the ATO/C system operates as a neural network with a learning function.

Claims

1. An automatic train operation apparatus incorporating a security function, for controlling a train operation, comprising:

an automatic train operation and control (ATO/C) system including a plurality of ATO/C units provided redundantly, each ATO/C unit having a fail safe configuration with a reliability ATO including a plurality of execution processors and a supervisory processor for monitoring operations of the execution processors and determining an output of said each ATO/C unit, all of the execution processors in the plurality of ATO/C units simultaneously executing an identical automatic train operation program for obtaining a control command output for controlling the train operation; and

majority logic means for selecting an output obtained by a majority of the ATO/C units of the ATO/C system as a control command output of the ATO/C system for controlling the train operation such that a total reliability R of the apparatus is greater than the reliability ATO of each ATO/C unit.

2. The automatic train operation apparatus of claim 1, wherein the supervisory processor of each ATO/C unit has a function of disconnecting said each ATO/C unit when a malfunction is detected in one of the execution processors in said each ATO/C unit.

3. The automatic train operation apparatus of claim 2, further comprising a continuation switch means for forcing the majority logic means to select an output obtained by a minority of the execution processors in the ATO/C units of the ATO/C system as the control command output of the ATO/C system at times when a majority of the ATO/C units in the ATO/C system are disconnected.

4. The automatic train operation apparatus of claim 1, wherein each execution processor operates as a neural network with a learning function.

5. An automatic train operation apparatus incorporating a security function, for controlling a train operation, comprising:

input means for entering external information signals;

an automatic train operation and control (ATO/C) system including at least one ATO/C unit having a fail safe configuration including a plurality of execution processors and a supervisory processor for monitoring normal operations of the execution processors and determining an output of said ATO/C unit, wherein the execution processors in the ATO/C unit simultaneously execute an identical automatic train operation program for obtaining a control command output for controlling the train operation by using the external information signals entered by the input means, such that automatic train operation for automatically controlling the train operation and automatic train control for providing a security function are integrally realized in each ATO/C unit; and

control means for controlling a power running notch and a brake notch of a train according to the control command output produced from the ATO/C unit.

6. The automatic train operation apparatus of claim 5, wherein the ATO/C system includes:

a plurality of ATO/C units, the ATO/C unit being redundantly provided;

a majority logic means for selecting an output obtained by a majority of the execution processors in the ATO/C units of the ATO/C system as the control command output to be outputted to the control means.

7. The automatic train operation apparatus of claim 6, wherein the supervisory processor of each ATO/C unit has a function of disconnecting said each ATO/C unit when a malfunction is detected in one of the execution processors in said each ATO/C unit.

8. The automatic train operation apparatus of claim 7, further comprising a continuation switch means for forcing the majority logic means to select an output obtained by a minority of the execution processors in the ATO/C units of the ATO/C system as the control command output form the ATO/C system, at times when a majority of the ATO/C units in the ATO/C system are disconnected.

9. The automatic train operation apparatus of claim 5, wherein each execution processor operates as a neural network with a learning function.

10. An automatic train operation method using a neural network with a learning function, comprising the steps of:

controlling a power running notch and a brake notch of a train by estimating a train speed, a running distance, and a stop position using the neural network with a learning function;

evaluating a quality of train operation in terms of factors including arrival time accuracy, a stop position accuracy, a comfortability, and an energy consumption efficiency, by using the neural network with a learning function in response to the controlling step; and

gradually adjusting the controlling step by using the learning function of the neural network in order to minimize any difference between the quality of the train operation evaluated by the evaluation step and a desired quality of the train operation.

11. The automatic train operation method of claim 10, further comprising the steps of:

dividing a total running distance of a train into a plurality of sections; and

performing the controlling step, the evaluating step, and the adjusting step at each of the plurality of sections.

12. The automatic train operation method of claim 10, wherein the quality at the evaluating step is evaluated by using an evaluation function including predetermined factors for determining the quality of train operation.

13. The automatic train operation method of claim 10, wherein the quality at the evaluating step is evaluated by comparing a running pattern curve resulting from the controlling step with a past superior running pattern curve.

14. The automatic train operation method of claim 13, wherein the past superior running pattern curve is updated every time a single cycle of the automatic train operation is completed.

15. The automatic train operation method of claim 10, wherein the quality at the evaluating step is evaluated by comparing a running pattern curve resulting from the controlling step with an ideal running pattern curve having a maximized factor for determining the quality.