JP4307394B2 - Multi-computer simulation device - Google Patents

Description

  The present invention relates to a multi-computer simulator device, such as a real-time traffic simulator, which performs a simulation on a desired event and displays the calculation result on a single large multi-screen composed of a plurality of displays.

  In a normal traffic simulator, for example, when a 10-minute traffic simulation is executed, a simulation (calculation) of the movement of a car for 10 minutes is performed in advance, and the result is then displayed on the display screen as an animation. Is called. In this case, it is just like watching a video image, and the same animation is displayed no matter how many times it is viewed later.

In contrast to such a normal traffic simulator, a real-time type traffic simulator displays a calculation result every moment on a display in real time.
In this method, calculation and animation display are performed at the same time. For example, the red / blue ratio (split) of a signal at an intersection is changed on the spot at any time, and how the traffic situation changes accordingly. There is an advantage that it can be observed.

  On the other hand, the area of traffic that can be handled by a single traffic simulation has expanded due to the recent improvement in computer computing power. At present, however, the display capability has not kept up.

Reference (1) (Shun-ichiro Nakamura et al. “Traffic Simulator NITTS and its Multi-processor” Information Processing Society of Japan 7-3, pp.15-12, November 2001)

As shown in the above-mentioned document, when a moving car is displayed on a display screen in a state where it can be identified, the range that can be displayed on one display is a block area of about 1000 m × 700 m.
Therefore, if a larger area is simulated and the entire wide area is displayed, the image of each vehicle becomes too small to be identified. If this is enlarged using the zoom function, each car can be identified, but if this is done, the image of the entire area cannot be grasped.

  In the real-time simulator as described above, it is extremely inconvenient to use such a zoom function because of its characteristics. That is, it goes without saying that it is preferable that each vehicle is always displayed in an identifiable state without using the zoom function. The above-mentioned document discloses a technique that solves this point by a multi-computer method.

Prior to disclosing the present invention, the prior art will be described in detail below with reference to the drawings in order to facilitate understanding of the present invention.
FIG. 1 is an explanatory diagram showing a configuration example of a conventionally known multi-monitor type simulator,
FIG. 2 is an explanatory diagram showing an example of the monitor screen of the simulator shown in FIG.
FIG. 3 is an explanatory diagram showing a breakdown of the calculation amount of the computer used in this simulator.
In the figure, reference numeral 1 denotes, for example, a computer equipped with an OS of Windows (registered trademark) 98 or later of Microsoft Corporation. The computer 1 has a multi-monitor function, and a plurality of (up to 10) displays can be connected to one computer so that a screen to be displayed can be divided and displayed.
The simulator shown in FIG. 1 uses this function to connect, for example, four displays 1-A, 1-B, 1-C and 1-D to one computer, One large display screen 1 that displays the simulation results of the entire region Z as a whole, displaying the simulation results in the regions A, B, C, and D adjacent to each other, as shown in FIG. -T is configured.
Thus, in the simulation screen shown in FIG. 2, 1-A, 1-B, 1-C, and 1-D are displays showing the simulation results in the aforementioned regions A, B, C, and D, respectively. Roads, 3 and 3 are city blocks, and 4 and 5 are vehicles traveling on the road.

Thus, if the area A, B, C and D is now 1000m x 700m, the total area Z is 4 times the area of 1000m x 700m, ie 2000m The area becomes × 1400m, and simulation results can be displayed there.
In this case, it can be seen that one large display 1-T having a display area four times that of one display is connected to the computer 1.

However, in this method, four displays are used and the display area is shared, but the display speed cannot be improved as compared with the case where one display is used.
The amount of computation required for the computer 1 is
a) Simulation for normal application program (hereinafter referred to as simulation calculation a);
b) Computation for converting the result of computation normally performed by the OS into image data displayed on the monitor and outputting it to the monitor (hereinafter referred to as display processing computation b).

As shown in FIG. 3, the computer 1 executes the simulation calculation a and the display processing calculation b for all the areas where the simulation is performed.
The calculation amount of the simulation calculation a is much smaller than the calculation amount required for the display processing calculation b.
The calculation load of the display processing calculation b is several to several tens of times higher than the calculation load of the simulation calculation a even with a simple model as shown in FIG.
Therefore, in this type of simulator, the display processing calculation b occupies most of the calculation amount.

  Therefore, in this method, a four-fold area simulation can be performed by four displays, but a problem that the real-time property is difficult to maintain, for example, occurs because the CPU is loaded four times.

In the current state of the art, if the calculation cycle of the simulation calculation is shortened, real-time display becomes difficult, so the cycle must be lengthened.
In the future, if the computation speed of the computer improves, it is possible to shorten the computation cycle and improve the display speed. However, under the same conditions, the display speed is higher than that of this conventionally known multi-display type simulator. If a method that can be realized is proposed, the invention will not lose its usefulness no matter how much the computing speed of the computer is improved.

  FIG. 4 is an explanatory view showing the configuration of a known multi-computer type simulator different from the above, FIG. 5 is an explanatory view showing a calculation area and a display area of a computer used in this simulator, and FIG. FIG. 7 is an explanatory diagram showing an operation load of each computer.

As shown in FIG. 4, this multi-computer type simulator is formed by connecting a central control device 10 such as an Ethernet (registered trademark) HUB and a plurality of personal computers 11, 12, 13 and 14 through a LAN cable 15. Yes, 11-A, 12-B, 13-C and 14-D are the displays of their personal computers 11, 12, 13 and 14, and the simulation results in the above-mentioned areas A, B, C and D, respectively. To display.
In FIG. 4, the displays 11-A, 12-B, 13-C and 14-D are shown in a line, but in reality, they are arranged in two rows and two stages as shown in FIG. And arranged so as to form one display screen as a whole.

  In the example of FIG. 3, the simulation area is divided into four, and the computer 11, 12, 13, and 14 are respectively subjected to the simulation calculation of the respective areas A, B, C, and D, and the respective displays 11-A, 12-B, and 13- C, 14-D performs display processing calculation to display the simulation calculation result, and each display 11-A, 12-B, 13-C, 14-D has a region A, B, C, D respectively. The display result is displayed.

FIG. 5 shows this, and the partial areas A, B, C, and D are combined to form the entire area, but each display is responsible for displaying which part in the entire area is surrounded by a thick line. The shaded area is shown.
For example, the simulation result of the shaded area A is displayed on the display 11-A, and the corresponding computer 11 is in charge of the simulation calculation of the shaded area A and also the area A. It shows that display calculation processing is performed. The same applies to other displays.

  At the start of the simulation, data is loaded into each computer, the area in charge is determined, and when the start button of one computer, for example, the computer 11, is pressed, it becomes the master computer, via the LAN 15 and the central controller 10. The master computer 11 issues a simulation start command to the other computers 12, 13, and 14, and all the computers 11, 12, 13, and 14 start simulation computations synchronously while communicating between the processors. The results are displayed on the screens of the respective displays 11-A, 12-B, 13-C, and 14-D.

That is, the four computers 11, 12, 13, and 14 perform simulation and display of each assigned area independently after starting the simulation operation at the same time.
As shown in FIG. 6, the image displayed on the display is the same as that shown in FIG.
The problem with this method is the boundary processing between computers.

In FIG. 6, four displays 11-A, 12-B, 13-C, and 14-D are arranged side by side, and simulation calculation results are displayed.
On each screen of the display 11-A, 12-B, 13-C, 14-D, the road 16 and the block 17 are drawn as the situation of the areas A, B, C, D, and the road 16 is between the screens. It is connected across.
18 and 19 are cars running on the road.

Looking at the problem of the boundary processing, for example, the case where the vehicle 18 moves from the area A of the computer 11 to the area B of another computer 12 becomes a problem.
Each computer can grasp all the situation in its own area, but cannot understand the situation of other computers.
For this reason, the speed at which the vehicle 18 should enter the region B of the computer 12 is unknown.

That is, if the road ahead is congested, the vehicle must enter at a low speed, and if the vehicle ahead is sufficiently free, the vehicle may enter at a free running speed.
For this reason, each computer sends the approach speed designation information to the adjacent computer at a predetermined time, for example, every second.

This specifies that each lane of the road connected between each other will run at a speed of several kilometers per hour. Vehicles 18 and 19 that are about to move to the computer 12 in FIG. 6 are driven to the speed specified with reference to this information, and prepare for the transition between computers.
This approach speed designation information is influenced not only by the traffic situation of area B but also by the position, number, speed, etc. of vehicles entering area B from area D, and is not related to the situation of area C. Absent.

In FIG. 6, a vehicle 19 is exactly the vehicle that is about to move to the region D of the computer 14. The next moment, the vehicle 18 disappears from the display area A shared by the computer 13 and appears in the shared area D of the computer 14.
For this purpose, vehicle transition information is sent between adjacent computers every predetermined time, for example, every 0.1 second.
This consists of information such as the presence / absence of a vehicle, the speed, the type of vehicle, etc., that move for each lane of the road connected to each other.

As described above, the conventional method requires boundary processing between computers and communication between computers.
FIG. 7 shows a computation load for such communication processing between computers. As described above, although depending on the situation, the calculation load is a calculation amount that is comparable to the display calculation process.
Furthermore, the problem with this method is that it is only an approximate method to compensate for the fact that the situation of the approach destination is not known at all.
Therefore, in such a system, the vehicle cannot be transferred between computers completely smoothly, that is, seamlessly (without connection).

  In other words, there is a problem that when the vehicle moves within the area of the same computer and when the vehicle moves in an area across the computers, the movement is not exactly the same. In addition, the two types of information described above must be regularly exchanged between the computers that share the areas adjacent to each other, and there is a problem in that it requires communication processing time between computers. It was.

The present invention has been made to solve such a problem, and an object of the present invention is to provide a simulation apparatus capable of greatly reducing the number of communication between computers and seamlessly connecting screens between the respective areas.
The present invention has been made with a focus on the fact that the computer calculation speed has been greatly improved by the development of computer technology in recent years, but the display capability has not caught up with it, and the problem of the present invention is that of simulation. An object of the present invention is to provide an apparatus capable of displaying a result on a large screen in real time.

In the above-described conventional apparatus, each computer calculates a partial area and displays the portion.
On the other hand, in the present invention, all the computers are allowed to perform simulation calculation for all areas, and the calculation results are displayed in the same manner as in the conventional apparatus. It is intended to be shared with.

Accordingly, communication such as vehicle movement information between computers becomes unnecessary, and at the same time, the vehicle can move seamlessly between computers.
Instead, each processor conventionally had to calculate only a partial area for the simulation operation, but in the present invention, it must be calculated for the entire area. In this respect, the load becomes large. Since the amount of computation itself is not so large, there is no problem in recent high-speed computers.

  In the following description of the present invention, in order to simplify the explanation, it is assumed that one display is connected to one computer. However, as described above, a plurality of displays are connected to one computer. It is clear that the present invention can be applied to cases.

Hereinafter, the present invention will be specifically described with reference to the drawings.
FIG. 8 is an explanatory diagram showing a configuration of a multi-computer type simulator according to the present invention, FIG. 9 is an explanatory diagram showing a calculation area and a display area of a computer used in the simulator, and FIG. 10 is an arithmetic load of each computer It is explanatory drawing which shows.
These devices and arrangements are the same as those shown in FIG. 4 described above, but are installed in the central control device 20 such as the Ethernet (registered trademark) HUB and the computers 21, 22, 23, 24. The software is quite different.
Also, in FIG. 8, the displays are arranged in a line, but in reality, they are arranged in two rows and two stages as shown in FIG. 9, and are arranged so as to form one display screen as a whole. It is.

In the present invention, at the start of the calculation, all the computers 21, 22, 23, 24 are loaded with the same data, started simultaneously, and first perform the same simulation calculation synchronously, Display processing calculation is performed for the area to which each is assigned.
This simulation calculation is performed for the entire region A, B, C, and D.
In addition, each of the computers 21, 22, 23, and 24 performs display processing calculations for the display areas A, B, C, and D, respectively, and displays the corresponding display devices 21-A, 22-B, 23-C, and 24, respectively. -D outputs data necessary for display.

FIG. 9 is an explanatory diagram showing the calculation sharing area of each of the computers 21, 22, 23, 24. Each computer covers all of the hatched areas, that is, the areas A, B, C, and D. Thus, a simulation calculation is performed, and a display calculation is performed for each assigned area surrounded by a thick frame.
Although the entire area is displayed in this way, according to the present invention, the simulation calculation is performed borderless regardless of the boundaries of the areas A, B, C, and D. The image passes through the boundary very smoothly.
More specifically, when the front of one vehicle leaves area A, passes through the boundary, and enters area B , the image of the front of the vehicle that disappears in area A appears in area B in real time. The vehicle image passes through the boundary smoothly and is displayed.

  For example, the computer 21 performs simulation calculation for all of the partial areas A, B, C, and D, but performs display calculation processing only for the portion of the partial area A. Since each computer performs the simulation calculation of the entire area in this way, communication between computers when a vehicle shifts between computers is not required as in the prior art. In addition, there is no awkwardness when the car moves between computers, and seamless screen connection can be realized.

Next, the operation of this apparatus will be described in more detail.
At the start of the simulation, data of the entire simulation area is loaded on each computer, and information on the display charge portion is set in each computer.
Each display device 21-A, 22-B, 23-C, 24-D of each computer 21, 22, 23, 24 displays a simulation start screen, and a start button is also displayed therein. .

When this start button is pressed with a mouse or the like from any of the four computers, that computer becomes the master computer and the others become slave computers.
The master computer instructs all slave computers to start the simulation all at once from the central controller 20 via the LAN 25, and itself starts the simulation operation.

In this way, the four computers start the simulation at the same time, but after that, each computer operates independently.
There may be some error between the internal clocks of each computer over a long period, but it can be said that there is no error in a short period such as during the simulation.
Therefore, after starting the simulation all at once, even if each moves independently, the real-time simulation is executed while maintaining the synchronization.

  Each computer performs simulation calculation for all areas including partial areas A, B, C, and D, but the display of the result is performed only for the partial area specified in advance, and the partial area is specified. Go on the display screen.

The calculation load of each computer at this time is as shown in FIG.
The blocks in the figure are the total calculation loads of the computers 21, 22, 23, and 24, respectively, and the simulation calculation processes over the entire simulation areas, that is, the areas A, B, C, and D, and the respective display persons. It shows that it consists of display processing calculation of the area.

The configuration of the present invention is not limited to the above-described embodiment. For example, an example in which four computers are used has been described above. In general, it is recommended to use m × n computers (m and n are positive integers) and arrange their displays in m rows and n stages. Is appropriately selected depending on the purpose of simulation.
Further, as a computer, a hybrid type using a plurality of multi-display types shown in FIG. 1 as a well-known example is also included in the present invention.
Although zoom technology is well known, it is not mentioned. However, when display of a partial area is designated by zoom designation, the partial area is divided into four parts, and each computer 21, 22, 23, 24 is assigned to each. It is clear that the display can be easily realized by those skilled in the art.
Furthermore, although an example in which Ethernet (registered trademark) HUB is used as a central control device for overall control of a plurality of computers has been described above, this is within the scope of the object of the present invention regardless of the name. It includes all devices having functions corresponding to this.

  Arithmetic processing such as communication between the computers is not necessary, the efficiency of the computation is increased, and there is no connection between display areas of each display, and seamless screen connection is possible.

  Since the present invention can be widely used not only for the above-mentioned traffic simulation but also for various event simulations, if this is implemented, there is a great benefit in a wide range of industrial fields.

An explanatory view showing a configuration example of a conventionally known multi-monitor type simulator, Explanatory drawing showing an example of the monitor screen of the simulator shown in FIG. Explanatory drawing which shows the breakdown of the computational amount of the computer used for said simulator, Explanatory drawing which shows the structure of the well-known multicomputer system simulator different from the above, Explanatory drawing showing the calculation area and display area of the computer used in this simulator, The figure which shows an example of the image displayed on each display, Explanatory drawing showing the calculation load of each computer of this conventional method, Explanatory drawing which shows the structure of the simulator which concerns on this invention, FIG. 8 is an explanatory diagram showing the calculation share of each computer in the simulator shown in FIG. FIG. 9 is an explanatory diagram showing a calculation load of each computer in the simulator shown in FIG.

Explanation of symbols

1, 11, 12, 13, 14: Computer
1-A, 1-B, 1-C, 1-D: Display device
11-A, 12-B, 13-C, 14-D: Display device
10, 20: Central control unit
15, 25: LAN cable
2, 16: road
3, 17: Block
4, 5, 18, 19: Vehicle
21, 22, 23, 24: Computer
21-A, 22-B, 23-C, 24-D: Display device A: Display area of computer 1 B: Display area of computer 2 C: Display area of computer 3 D: Display area of computer 4

Claims (5)

A plurality of computers (21, 22, 23, 24) each having at least one display (21-A, 22-B, 23-C, 24-D), and a central control unit that controls the plurality of computers And a control device (20) for simulating an event occurring in a predetermined display area, the displays (21-A, 22-B, 23-C, 24-D) of the display area is arranged so as to constitute a single screen displaying, the predetermined display area is divided into a plurality of computers (21, 22, 23, 24) each display coverage areas (a, B, C, D) in Assigned to each computer (21, 22, 23, 24), and the computer (21, 22, 23, 24) performs a simulation calculation for the above event and a display process for each display area (A, B, C, D). Calculations are performed and the results are displayed on the corresponding displays. In a multi-computer simulator that is displayed on a ray (21-A, 22-B, 23-C, 24-D), and thus displays all of the simulation results in the predetermined display area ,
The multi-computer type simulator described above, wherein all the computers (21, 22, 33, 24) are configured to perform the same simulation calculation processing in synchronism with respect to all display areas.   2. The multicomputer simulator according to claim 1, wherein the central control device is an Ethernet (registered trademark) HUB.   The multicomputer simulator according to claim 1 or 2, wherein each of the computers (21, 22, 23, 24) has one display (21-A, 22-B, 23-C, 24-D). .   4. The multi-computer type simulator according to claim 3, wherein the computer is provided with n × m (n and m are positive integers) units, and the display is arranged for n units and arranged in m stages.   The multi-computer type simulator according to claim 1 or 2, wherein each of the computers (21, 22, 23, 24) has a plurality of displays.

Images