JPS63228020A - Sensor bus system - Google Patents

Abstract

PURPOSE:To maintain high-reliability data output performance by the arrangement of the irreducible number of sensors by coupling the respective sensors by a data bus, and processing information from the data bus by a sensor data control computer. CONSTITUTION:All flight control and navigation sensors such as a rate gyro sensor 10, an accelerometer 11, an attitude angle sensor 12, an airspeed meter 13, a static pressure gauge 14, an elevation angle meter 15, and a lateral slide meter 16 are coupled by the dedicated data bus 17, whose information is inputted to the sensor data control computer 18. The computer 18 performs processes such as total analytic redundancy calculation and redundancy control among different kinds of sensor information. Then the output of the computer 18 is supplied to a flight control computer, a flight control computer, etc., through an external bus for external utilization.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention is a sensor applied to a moving object sensor such as an aircraft.
Concerning the pass system.

[Conventional technology]

Conventionally, in aircraft sensor systems, each sensor is handled individually. Each sensor is designed to output data that is easy to handle by the central digital control computer, and some of the sensors are partially calibrated digital data that is strap-down (directly attached to the aircraft) and subjected to calculation correction before being output. It has been commercialized.

For example, in conventional gyros and acceleration sensors, 1 ()
When high reliability such as -2o failure time or more is required, a method of reducing the number of sensors by diagonally arranging them in a regular dodecahedral arrangement as shown in FIG. 4 has been considered.

This is Strike 2. This includes f-down technology.

Fig. 5 shows an example of a conventional sensor system. moves, the aircraft dynamics, and Kusu 2 occurs. Its output is accelerometer < nz) s a angle of attack meter (α)4. It is detected by rate/gyro sensor (q) 5, etc. These are normal FDI (Failure Det@ction a
nd l5olation (majority voting principle) 8, the signals are selected and used, for example, in a flight control computer. Now, if one of the rate gyros is broken and two are operating, it is normal to know when the failure occurred, but it is not possible to identify which rate gyro is in failure. In this case, accelerometer 3. The signal from the angle of attack meter 4 is input to the Kalman filter/aircraft motion model 6 to estimate q, and is compared with the output of the actual rate gyro sensor using analytical FDI 7 to determine the correct rate and gyro. , select that output. Kafman filter/
The aircraft motion model 6 and the analytical FDI 7 are called sensor observers (b), and are used to assist in determining failures of actual instruments.

In this way, in conventional sensor systems, an analysis redundant system that requires information from other sensors is considered in order to detect and identify failures, etc. This is the concept of individual sensors that produce output.

FIG. 6 shows another example of a conventional sensor system, in which the movement of the actuator 1 is transmitted throughout the airframe mechanics 2, an accelerometer (nz), 3e an angle of attack meter (α), 4. Rate J.R.
It is output to the flight control sensor (q) 5, and is input to the Kalman filter/aircraft motion model 6 and control law 9 to configure the flight control law ff. Now rate gyro sensor (
q) If 5 stops working at all, the accelerometer (n'z
) 3 e Angle of attack meter (α) 4 only Kalman dispute filter/
Based on the input to the aircraft motion model 6, the force y-man φ filter/aircraft motion model 6 estimates the rate signal and applies the control law 9.
Enter. Here, the Kalman filter/aircraft motion model 6 and control law 9 are functions built into the flight control computer (c).

In this way, at the aircraft level, there is a concept of analysis redundancy that uses the kinematic model to eliminate sensors that are no longer receiving information and estimate necessary information. There is.

[Problem that the invention seeks to solve]

Conventional sensor systems have the following drawbacks.

■ Greater sensor redundancy when reliable sensor data is required. That is, if the sensor alone outputs highly reliable data, six sensors are required as shown in FIG. 4 even if all the latest methods are applied.

■ Redundancy is high because analysis redundancy, which requires all other sensor information, is not considered for obliquely arranged sensors.

■ Since analysis redundancy is considered individually at each sensor level, the computational load for each sensor is high.

■ Redundancy management between each sensor and different types of sensors is performed by a central flight control computer, which requires a large calculation load.

■ In the case of (■) above, even if intelligent sensors are placed locally, the data bus connecting them will not have enough capacity to handle only the sensors.

[Means for solving problems]

In order to solve the above problems, the present invention connects each sensor with a data bus, processes the information on this data bus in a sensor data management converter, and sends the output from this sensor data management computer to an external path. It is intended to be supplied to

[For production]

As in the above method, a sensor/data management combination is installed to process all the information on the data bus connecting each sensor, and in order to perform comprehensive analysis redundant calculations and redundant management between different sensor information, each With this sensor, highly reliable data output performance can be maintained with a minimum number of sensors.

[Example]゛ Embodiments of the present invention will be described in detail below with reference to the drawings.

FIG. 1 shows an embodiment of the present invention, in which a rate gyro sensor 10. Accelerometer 11. Attitude angle sensor 12, airspeed indicator 13. Static pressure gauge14. All flight control and navigation sensors such as the angle of attack meter (α) 15 and the horizontal pessimometer (β) 16 are connected by a dedicated data node 17. Sensor and data management combination, −
Entered on evening 18th. This sensor data management computer 18 performs processing such as comprehensive analysis redundancy calculation and redundancy management between different types of sensor information. The output from the sensor data management computer I8 is supplied to the flight control computer, flight management computer, etc. via an external path for external use.

FIG. 2 shows the contents of the sensor data management computer 18. That is, the output of each sensor 19 (majority level signal selection) is input to the redundancy check unit 20 and the remaining sensor redundancy is checked. If the redundancy is 3 or more, it is directly sent to the various sensor data buffers 24. Output to. Redundancy is 2! Or, in the case of one layer, it is input to the fault identification circuit (Kalman filter/aircraft motion model) 21, but the function of the fault identification circuit 21 is the output of the fault identification circuit 21, which performs the fault sensor identification shown in FIG. is output to various sensor data buffers 24 via the sensor failure check section 22t. If the redundancy checking unit 20 has one redundancy, the failure identification circuit 21 and the sensor failure checking unit 22
If the sensor is determined to be malfunctioning, the fact that all of the sensor information is lost is input to the sensor information generation circuit (Kalman filter/aircraft motion model) 23.

The function of the sensor information generation circuit 23 is the information generation function shown in FIG. 6, and its output is output to various sensor data buffers 24.

Figure 3 is a specific example of the present invention, showing a sensor path system to which the concepts of Figures 1 and 2 are applied.When reliability of 10 failures/hour or more is required as a sensor path system. Show about. In other words, the rate gyro sensor IQ is arranged in a skewed manner and normally consists of 6 regular dodecahedrons as shown in Fig. 4. Since this can be performed by the sensor data management computer 18, the five cone arrangement configuration is adopted. In addition, the accelerometer 11 is normally composed of six pieces, but the sensor data management computer 18
Since it is possible to generate acceleration information using the information from the gyro sensor 10, the number is further reduced to four regular tetrahedrons. Furthermore,
Angle of attack meter (α) 15, horizontal angle meter (β) 16. For sensors such as barometric altimeter 25° and static pressure gauge 14, a normal high-reliability system would have a 3- to 4-layer configuration, but by using a data creation function that uses information from other sensors, it is possible to
It has a heavy structure.

Second, when compared to conventional sensor systems, conventional sensor systems have increased sensor redundancy when highly reliable sensor data is required. That is, if the sensor alone outputs highly reliable sensor data, six sensors are required as shown in FIG. 4 even if the latest method is applied. On the other hand, in this embodiment, if analytical redundancy is used for fault discovery and identification, a configuration of five 6-1 stacks may be sufficient, and if used for sensor data creation, the configuration may be reduced by one to four. However, in the case of 4 sensors, other sensor information is active and the necessary data needs to be created from it, so the number of core sensors (in this case, the rate gyro sensor) needs to be kept at the level of 5.

Furthermore, in this embodiment, if the concepts of FIGS. 2 and 3 were broadly applied to all sensor systems, processing of sensor data would become somewhat complicated, so a dedicated path and computer were used. As a result, all sensor information is output from one dart and centrally managed, making it convenient to understand its status.

〔Effect of the invention〕

As described above, according to the present invention, a sensor data management computer is installed to perform comprehensive analysis redundant calculations and redundant management between different types of sensor information, so that sensor data can be centrally managed and the output window can be I can make it into a book. Therefore, in each sensor system, the number of constituent sensors and the calculation load can be reduced, so that each sensor can maintain highly reliable data output performance with a minimum number of sensors arranged.

[Brief explanation of drawings]

FIG. 1 is a configuration explanatory diagram showing one embodiment of the present invention, FIG. 2 is a configuration explanatory diagram showing an example of the sensor data management computer of FIG. 1, and FIG. 3 is a configuration explanatory diagram showing a specific example of the present invention. 4 to 6 are configuration explanatory diagrams showing conventional sensor systems, respectively. 1'0... Rate gyro sensor, 11... Accelerometer, 12... Attitude angle sensor, 13... Air speed meter, 14... Static pressure gauge, 15... Angle of attack meter, 16... Lateral slip meter, 12... Data bus, 18... Sensor 9 data management computer. Figure 4 Figure 5 Figure 6E

Claims (1)

[Claims] A sensor bus characterized in that each sensor is connected by a data bus, information on this data bus is processed by a sensor data management computer, and output from this sensor data management computer is supplied to an external bus. system.