Driving simulation system

Abstract

To more properly simulate vehicle behavior.SOLUTION: The driving simulation system includes a cockpit which has an operation device necessary for driving a vehicle to be simulated and on which an evaluator can ride, a rocking device for rocking the cockpit, and a control device for controlling the rocking device based on an operation of the operation device by the evaluator riding on the cockpit. A vehicle characteristic to be evaluated by an evaluator is predicted using a machine learning model in which a shape of a traveling course displayed at least in front of a cockpit and operation information of an operation device of the cockpit are input variables and a vehicle characteristic evaluated by the evaluator is an output variable, the displayed shape of the traveling course, and the operation information, and a swing device is controlled based on the predicted vehicle characteristic.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present disclosure relates to a driving simulation system. [Background technology]

[0002] Conventionally, a driving simulation system of this type has been proposed that includes a cockpit (vehicle interior) in which a driver sits and that has operating devices necessary for driving the vehicle to be simulated, a rocking device for rocking the cockpit, and a control device (control computer) that controls the rocking device based on the operation of the operating device in the cockpit (see, for example, Patent Document 1). In this system, the driver selects one simulation environment from multiple simulation environments. Then, the system changes a motion calculation model of the vehicle to be simulated according to the simulation environment selected by the driver, and controls the rocking device based on the motion calculation model. This makes it possible to simulate the behavior of the vehicle in the environment selected by the driver. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2003-316249 Summary of the Invention [Problem to be solved by the invention]

[0004] However, the above-mentioned driving simulation system can only simulate the behavior of a vehicle in a specific simulation environment. When the above-mentioned driving simulation system is used by an evaluator in a cockpit to evaluate vehicle characteristics, it is desirable to simulate vehicle behavior that allows the evaluator to accurately evaluate the vehicle characteristics.

[0005] The driving simulation system of the present disclosure has a primary objective of more accurately simulating vehicle behavior. [Means for solving the problem]

[0006] The driving simulation system of the present disclosure employs the following measures to achieve the above-mentioned main object.

[0007] The driving simulation system of the present disclosure includes: a cockpit that has the operating devices necessary for driving the simulated vehicle and that an evaluator can sit in; A rocking device that rocks the cockpit; a control device that controls the rocking device based on an operation of the operating device by the evaluator sitting in the cockpit; A driving simulation system that allows the evaluator sitting in the cockpit to experience vehicle behavior when driving the vehicle, comprising: The control device predicts the vehicle characteristics to be evaluated by the evaluator using a machine learning model that uses, as input variables, the shape of the traveling course displayed at least in front of the cockpit and operation information of the operation device in the cockpit and the vehicle characteristics evaluated by the evaluator as output variables, the shape of the traveling course that is displayed, and the operation information, and controls the rocking device based on the predicted vehicle characteristics. The gist of this is as follows.

[0008] The driving simulation system disclosed herein uses a machine learning model that uses the shape of the driving course displayed at least in front of the cockpit and operation information of the cockpit operation device as input variables and vehicle characteristics evaluated by an evaluator as output variables, and predicts the vehicle characteristics evaluated by the evaluator using the displayed shape of the driving course and the operation information, and controls the rocking device based on the predicted vehicle characteristics, thereby enabling more accurate simulation of vehicle behavior. [Brief explanation of the drawings]

[0009] [Figure 1] 1 is a diagram showing an outline of the configuration of a driving simulation system 10 according to the present embodiment. [Figure 2] FIG. 1 is an explanatory diagram for explaining a method for creating a machine learning model. [Figure 3] FIG. 2 is an explanatory diagram for explaining vehicle characteristics evaluated by an evaluator. DETAILED DESCRIPTION OF THE INVENTION

[0010] An embodiment of the present disclosure will be described with reference to the drawings. Fig. 1 is a diagram showing an outline of the configuration of a driving simulation system 10 of this embodiment. The driving simulation system 10 of this embodiment is used to evaluate vehicle characteristics and includes a cockpit 20, a rocking device 30, a projector 32, a screen 34, and a control device 40.

[0011] The cockpit 20 is configured to simulate the interior space of the vehicle whose characteristics are to be evaluated, and is equipped with a steering wheel 22, an accelerator pedal 24, a brake pedal 26, as well as operating devices necessary for driving the vehicle, such as a shift lever, and a seat for the evaluator.

[0012] The rocking device 30 supports the cockpit 20 so that it can rock, and is configured as a six-axis motion base that enables the cockpit 20 to move in the front-to-back, left-to-right, and up-and-down directions, as well as to tilt left and right (roll), tilt front-to-back (pitch), and rotate left and right (yaw) by extending and retracting six electric cylinders. The rocking device 30 is controlled by a control device 40.

[0013] The projector 32 projects an image based on image data of the driving course from the viewpoint of the driver's seat onto a screen 34 that is arranged to surround the cockpit 20 over a 270-degree angle, including the front of the cockpit 20. The projector 32 is controlled by a control device 40.

[0014] The control device 40 includes a CPU 40a, a ROM 40b, a RAM 40c, a microcomputer having a flash memory, input / output ports, and communication ports (not shown), various drive circuits, and various logic ICs. The control device 40 receives signals from various sensors in the cockpit 20. Examples of the various sensors include a steering angle θst from a steering angle sensor 22a that detects the steering angle of the steering wheel 22, an accelerator operation amount AP from an accelerator position sensor 24a that detects the depression amount of the accelerator pedal 24, and a brake operation amount BP from a brake pedal position sensor 26a that detects the depression amount of the brake pedal 26. The control device 40 outputs various signals. Examples of the output signals include an audio output signal to a speaker 28 in the cockpit 20, a video output signal to a projector 32, and a control signal to the rocking device 30.

[0015] The control device 40 stores in ROM 40b the specifications of the vehicle to be evaluated (such as the vehicle weight and the main power source for driving the vehicle, such as the engine or motor) and image data of the driving course to be projected onto the screen 34. Examples of the driving course include an actual urban area or a test circuit. The control device 40 stores a machine learning model created in advance in ROM 40b. Here, a method for creating the machine learning model will be described.

[0016] FIG. 2 is an explanatory diagram illustrating a method for creating a machine learning model. FIG. 3 is an explanatory diagram illustrating vehicle characteristics evaluated by an evaluator. At point P1 shown in FIG. 3, the course shape of the driving course heading to point P2 is a straight line, and the evaluator in cockpit 20 evaluates the vehicle's "straight-line performance (e.g., whether it is stable)" as a vehicle characteristic based on the bodily sensation inside the cockpit 20. At point P2 shown in FIG. 3, the course shape of the driving course heading to point P2 is a curved shape, and the evaluator evaluates the vehicle's "curve-entry performance" as a vehicle characteristic based on the bodily sensation inside the cockpit 20. At point P3 shown in FIG. 3, the course shape heading to point P4 is a curved shape, and the evaluator evaluates the vehicle's "curve-turning performance" as a vehicle characteristic based on the bodily sensation inside the cockpit 20. At point P4 in FIG. 3 , the shape of the course heading toward point P5 is such that it departs from a curve. The evaluator evaluates the vehicle's "curve exit performance" as a characteristic of the vehicle based on the sensations experienced in the cockpit 20. Thus, there is a correlation between the shape of the course and the vehicle characteristic evaluated by the evaluator. For example, if the evaluator were to evaluate the vehicle's "curve entry performance" as a characteristic of the vehicle at point P2, the evaluator would operate the steering wheel 22, accelerator pedal 24, and brake pedal 26 according to the simulated vehicle speed Vsim of the simulated vehicle so that the vehicle is positioned on the outer diameter side of the course before entering the curve. The simulated vehicle speed Vsim is calculated as the current value of the simulated vehicle based on the accelerator operation amount AP, the brake operation amount BP, and vehicle specifications. Thus, there is a correlation between the operation information in the cockpit 20, such as the operation amounts of the steering wheel 22, accelerator pedal 24, and brake pedal 26, the simulated vehicle speed Vsim, and the vehicle characteristic evaluated by the evaluator. Taking these factors into consideration, the control device 40 creates a machine learning model by machine learning, with the course shape up to the next point for each predetermined point, the operation information of the cockpit 20, and the simulated vehicle speed Vsim simulated at that point as input variables, and the vehicle characteristics evaluated by the evaluator as output variables.

[0017] In the driving simulation system 10 configured as described above, when an evaluator enters the cockpit 20 and performs a predetermined simulated driving start operation, the control device 40 repeatedly executes simulation control every predetermined time Δt (e.g., every few msec). In the simulation control, the control device calculates the vehicle position Pv on the driving course based on the accelerator operation amount AP, the brake operation amount BP, the steering angle θst, and the simulated vehicle speed Vsim, and controls the projector 32 to display an image of the driving course from the viewpoint of the driver's seat at the vehicle position Pv on the screen 34 from the ROM 40b. The control device 40 then predicts the vehicle characteristics to be evaluated by the evaluator after the predetermined time Δt using the course image at the vehicle position Pv, ​​the accelerator operation amount AP, the brake operation amount BP, the steering angle θst, the simulated vehicle speed Vsim, and the created machine learning model. Furthermore, the control device 40 controls the rocking device 30 so that the longitudinal displacement, lateral displacement, vertical displacement, roll angle, pitch angle, and yaw angle of the cockpit 20 are appropriate for the predicted vehicle characteristics. For example, when the evaluator predicts that the "curve entry performance" will be evaluated after a predetermined time Δt, the cockpit 20 adjusts the longitudinal displacement, lateral displacement, vertical displacement, roll angle, pitch angle, and yaw angle so that the cockpit 20 will exhibit the vehicle behavior when entering a curve after the predetermined time Δt. This control allows the vehicle behavior to be more accurately simulated, improving the evaluator's evaluation of the vehicle's characteristics.

[0018] According to the driving simulation system 10 of this embodiment described above, a machine learning model is used in which the shape of the driving course displayed at least in front of the cockpit 20, the accelerator operation amount AP, the brake operation amount BP, the steering angle θst, and the simulated vehicle speed Vsim are used as input variables, and the vehicle characteristics evaluated by the evaluator are used as output variables.The shape of the displayed driving course, the accelerator operation amount AP, the brake operation amount BP, the steering angle θst, and the simulated vehicle speed Vsim are used to predict the vehicle characteristics to be evaluated by the evaluator, and the rocking device 30 is controlled based on the predicted vehicle characteristics, thereby making it possible to more accurately simulate vehicle behavior.

[0019] In the above-described embodiment, the machine learning model is created using the shape of the travel course, the accelerator operation amount AP, the brake operation amount BP, the steering angle θst, and the simulated vehicle speed Vsim as input variables, but at least one of the accelerator operation amount AP, the brake operation amount BP, and the steering angle θst may be used, or operation information of an operation device other than the accelerator operation amount AP, the brake operation amount BP, and the steering angle θst may be used as an input variable. Also, the simulated vehicle speed Vsim does not have to be used as an input variable, and a parameter indicating a vehicle state other than the simulated vehicle speed Vsim, such as the acceleration of the vehicle that is simulated together with the simulated vehicle speed Vsim, may be used as an input variable.

[0020] The correspondence between the main elements of the embodiment and the main elements of the invention described in the "Means for Solving the Problem" section will be explained. In the embodiment, the cockpit 20 corresponds to the "cockpit," the rocking device 30 corresponds to the "rocking device," and the control device 40 corresponds to the "control device."

[0021] The correspondence between the main elements of the embodiments and the main elements of the invention described in the "Means for Solving the Problem" section does not limit the elements of the invention described in the "Means for Solving the Problem" section, since the embodiments are examples for specifically explaining the mode for implementing the invention described in the "Means for Solving the Problem" section. In other words, the interpretation of the invention described in the "Means for Solving the Problem" section should be based on the description in that section, and the embodiments are merely specific examples of the invention described in the "Means for Solving the Problem" section.

[0022] The above describes embodiments for implementing the present disclosure, but the present disclosure is not limited to these embodiments and can, of course, be implemented in various forms within the scope that does not deviate from the gist of the present disclosure. [Industrial Applicability]

[0023] The present disclosure can be used in a driving simulation system. [Explanation of symbols]

[0024] 10 driving simulation system, 20 cockpit, 22 steering, 22a steering angle sensor, 24 accelerator pedal, 24a accelerator position sensor, 26a brake pedal position sensor, 26 brake pedal, 28 speaker, 30 oscillation device, 32 projector, 34 screen, 40 control device, 40a CPU, 40b ROM, 40c RAM, P1 to P4 points.

Claims

[Claim 1] a cockpit that has the operating devices necessary for driving the simulated vehicle and that an evaluator can sit in; A rocking device that rocks the cockpit; a control device that controls the rocking device based on an operation of the operating device by the evaluator sitting in the cockpit; A driving simulation system that allows the evaluator sitting in the cockpit to experience vehicle behavior when driving the vehicle, comprising: The control device predicts the vehicle characteristics to be evaluated by the evaluator using a machine learning model that uses, as input variables, the shape of the traveling course displayed at least in front of the cockpit and operation information of the operation device in the cockpit and the vehicle characteristics evaluated by the evaluator as output variables, the shape of the traveling course that is displayed, and the operation information, and controls the rocking device based on the predicted vehicle characteristics. Driving simulation system.

Images