JP7117376B2 - Electronic control unit evaluation device - Google Patents

Description

The present invention relates to an evaluation device for electronic control units.

In recent years, in order to realize a safe and comfortable automobile society, research on driving support systems and automatic driving systems has been active. In the future, as the level of automated driving improves, it will be necessary to develop complex control logic that includes the surrounding environment, such as merging on expressways and turning left or right at intersections. Regarding the verification evaluation of vehicle elements, it is expected that a huge number of test cases will be required because the control logic will become more complicated as the level of automated driving improves.

The current verification evaluation is centered on actual vehicles (referred to as actual vehicles). However, in the future, the number of man-hours required for testing will become even more enormous, and it will be difficult to perform verification evaluations only with tests using actual vehicles. Therefore, evaluation methods that utilize simulations such as HILS (Hardware In the Loop Simulator) will become commonplace. is coming.

HILS is a virtual environment in which an actual control system is constructed on a computer by constructing an evaluation device by combining real objects for control objects to be evaluated, modeling other vehicle elements, and constructing evaluation equipment. It is a tool to evaluate in As for HILS, for example, the technique disclosed in Patent Document 1 is known.

JP 2010-26845 A

Here, when constructing a HILS whose controlled object is a vehicle element that calculates various values based on various sensor information, it is necessary to provide the HILS with sensor information at the timing requested by the controlled object. For example, if the object to be controlled is an in-vehicle camera, it is necessary to input an image to the in-vehicle camera at the timing when the on-vehicle camera releases the shutter. Since the HILS cannot input an actual image to the in-vehicle camera, a virtual image created by CG (Computer Graphics) is input.

The technique of Patent Document 1 does not have a function of feeding back the shutter signal output by the camera control ECU. Therefore, if the drawn CG is input to the camera control ECU at the timing when the CG is drawn, the CG is input to the camera control ECU with a delayed timing, and the in-vehicle camera cannot perform correct recognition and control. .

SUMMARY OF THE INVENTION It is an object of the present invention to provide a technology capable of appropriately testing the operation of an electronic control unit.

In order to solve the above-described problems, an evaluation apparatus for an electronic control unit according to the present invention provides computer graphics (CG) from a viewpoint of a virtual camera mounted on a virtual vehicle running on a virtual road built on a computer. an evaluation device for testing the operation of the electronic control unit by inputting into the vehicle behavior simulator for calculating the behavior information of the virtual vehicle, the CG request signal output from the electronic control unit, and the a camera viewpoint specifying function for calculating viewpoint information of the virtual camera based on the behavior information of the virtual vehicle; a request signal of the CG output from the electronic control unit; and the viewpoint information of the virtual camera. , a CG generation function for generating the CG with the virtual camera as a viewpoint; and a CG change function for converting the generated CG into a format that can be input to the electronic control unit.

According to the present invention, the operation of the electronic control unit can be properly tested.

The block diagram of HILS which controls ECU. FIG. 4 is a configuration diagram of a camera viewpoint specifying function; 4 is a conceptual diagram of a camera viewpoint specifying function; FIG. A flowchart of a camera viewpoint specifying function. FIG. 4 is a configuration diagram of a left CG generation function and a right CG generation function; Flowchart of left CG generation function and right CG generation function. The block diagram of a CG conversion function. The block diagram of ECU.

Several embodiments will be described in detail with reference to the drawings. It should be noted that the examples described below do not limit the invention according to the claims, and that all of the elements and combinations thereof described in the examples are essential to the solution of the invention. is not limited.

FIG. 1 is a configuration diagram of a HILS that controls an ECU.

A HILS (Hardware In the Loop Simulator) as an example of an "evaluation device" is a device for testing the operation of an ECU (Electronic Control Unit) 6 as an example of an "electronic control unit". CG (Computer Graphics) is input to the ECU 6 with a stereo camera as an example of a "virtual camera" mounted on a virtual vehicle running on a virtual road constructed on a computer. A stereo camera images the front of the virtual vehicle.

HILS includes a vehicle behavior simulator 1 , a camera viewpoint specifying function 2 , a left CG generation function 3 , a right CG generation function 4 and a CG conversion function 5 .

A camera output data format S6, which will be described later, is input to the ECU 6 from the CG conversion function 5, and vehicle information S11 such as vehicle speed is input from the vehicle behavior simulator 1. FIG. The ECU 6 outputs to the vehicle behavior simulator 1 requests for an accelerator signal S8, a brake signal S9, and a steering wheel operation signal S10. Further, the ECU 6 sends a shutter signal S7, which is an example of a "request signal" requesting CG from a stereo camera mounted on a virtual vehicle traveling on a virtual road built on a computer, to the camera viewpoint specifying function 2. , the left CG generation function 3, and the right CG generation function 4, respectively.

The vehicle behavior simulator 1 calculates behavior information of a virtual vehicle running on a virtual road built on a computer at 1 ms intervals. The behavior information of the virtual vehicle may include the coordinates and posture S1 of the virtual vehicle in the virtual space (simulation space). As a result, the position and orientation of the virtual vehicle can be properly grasped according to the running state.

When calculating the behavior of the virtual vehicle, the vehicle behavior simulator 1 performs calculation according to the accelerator signal S8, the brake signal S9, and the steering wheel operation signal S10 input from the ECU 6 in addition to the information operated by the driver. The vehicle behavior simulator 1 transmits the calculated coordinates and posture S1 of the virtual vehicle to the camera viewpoint specifying function 2 . Furthermore, the vehicle behavior simulator 1 transmits vehicle information S11 such as vehicle speed to the ECU 6 .

The camera viewpoint specifying function 2 receives the coordinates and attitude S1 of the virtual vehicle from the vehicle behavior simulator 1, and receives a shutter signal S7 as an example of a "request signal" from the ECU 6. FIG. The camera viewpoint specifying function 2 calculates the viewpoint information of the left and right virtual cameras at the timing when the ECU 6 requests CG based on the coordinates and posture S1 of the virtual vehicle and the shutter signal S7. The viewpoint information of the left and right virtual cameras may include the mounting angle and attitude of the virtual camera with respect to the virtual vehicle. As a result, not only the running state of the virtual vehicle but also the direction of the virtual camera can be properly grasped.

Specifically, the camera viewpoint specifying function 2 specifies the coordinates and the attitude S1 of the virtual vehicle when the shutter signal S7 is received, based on the coordinates and the attitude S1 of the virtual vehicle. Then, the camera viewpoint specifying function 2 calculates left viewpoint coordinates and posture S2 and right viewpoint coordinates and posture S3 based on the specified coordinates and posture S1 of the virtual vehicle.

The left CG generation function 3 receives the left viewpoint coordinates and the attitude S2 from the camera viewpoint specifying function 2 and receives the shutter signal S7 from the ECU 6 . The left CG generation function 3 generates the left CGS 4 when the shutter signal S7 is received at the timing requested by the ECU 6 based on the left viewpoint coordinates and attitude S2 and the shutter signal S7. The left CG generation function 3 outputs the generated left CGS 4 to the CG conversion function 5 .

The right CG generation function 4 receives the coordinates of the right viewpoint and the orientation S3 from the camera viewpoint specifying function 2 and receives the shutter signal S7 from the ECU 6 . The right CG generation function 4 generates the right CGS 5 upon receipt of the shutter signal S7 at the timing requested by the ECU 6, based on the right viewpoint coordinates and orientation S3 and the shutter signal S7. The right CG generation function 4 outputs the generated right CGS 5 to the CG conversion function 5 .

The CG conversion function 5 receives the left CGS4 from the left CG generation function 3 and the right CGS5 from the right CG generation function 4 . The CG conversion function 5 converts the left CGS4 and the right CGS5 into a camera output data format S6, which is a format that the ECU 6 can receive.

FIG. 2 is a configuration diagram of the camera viewpoint specifying function.

The camera viewpoint identification function 2 includes a behavior identification unit 201 , a left viewpoint calculation unit 202 and a right viewpoint calculation unit 203 .

The behavior specifying unit 201 specifies the coordinates and the posture S1 of the virtual vehicle used for viewpoint calculation of the stereo camera based on the shutter signal S7 among the coordinates and the posture S1 of the virtual vehicle updated at intervals of 1 ms. Based on the specified coordinates and orientation S1 of the virtual vehicle, the left viewpoint calculation unit 202 calculates the left viewpoint coordinates and the viewpoint S2. Similarly, the right viewpoint calculation unit 203 calculates right viewpoint coordinates and a viewpoint S3 based on the specified coordinates and orientation S1 of the virtual vehicle.

FIG. 3 is a conceptual diagram of the camera viewpoint specifying function. FIG. 4 is a flowchart of the camera viewpoint specifying function.

A stereo camera is mounted inside the virtual vehicle, as shown in FIG. The stereo camera may be installed in front of the room mirror inside the virtual vehicle, with the left and right virtual cameras lined up side by side, for example.

The vehicle simulator 1 performs behavior calculations Tn−1, Tn, and Tn+1 of the virtual vehicle at intervals of 1 ms (S402, S403, and S404), and outputs the calculation results to the behavior identification unit 201. FIG. Here, when the vehicle simulator 1 independently calculates the coordinates and posture S1 of the virtual vehicle for each of the left and right stereo cameras and inputs the calculation results to the behavior identification unit 201, the coordinates used for the left viewpoint and the It may be different from the coordinates used in the viewpoint. In particular, when stereo cameras are used, the parallax of the two cameras is used to calculate the distance to the target object, so viewpoint information with different CG drawing (shutter) timings with the left and right virtual cameras as viewpoints does not work as a simulator. .

Therefore, the shutter signal S7 output from the ECU 6 is used as a signal specifying the coordinates and orientations S2 and S3 of the left and right virtual cameras. Until the shutter signal S7 is input, the behavior identifying unit 201 holds the coordinates and attitude S1 of the virtual vehicle as data Tn-1 (S405). Update as data Tn, Tn+1, . . . (S406, S408). When the ECU 6 outputs a shutter signal S7 that is timing for CG synchronization (S401), the behavior identifying unit 201 obtains the held data Tn (S407), and transfers the obtained data Tn to the left viewpoint calculation unit 202. , to the right viewpoint calculation unit 203 . Thus, by specifying the coordinates and posture S1 of the virtual vehicle, the left viewpoint calculation unit 202 and the right viewpoint calculation unit 203 synchronize the left viewpoint coordinates and viewpoint S2 with the right viewpoint coordinates and viewpoint S3. (S409, S410), and the CG with the left and right virtual cameras as viewpoints can be synchronized. Therefore, the left and right virtual cameras are prevented from being calculated with different virtual vehicle coordinates and attitude S1.

When the left viewpoint calculation unit 202 receives the coordinates and orientation S1 of the virtual vehicle from the behavior identification unit 201, the left viewpoint calculation unit 202 calculates the mounting position of the left virtual camera and the mounting orientation of the yaw angle, roll angle, and pitch angle. Then, the left viewpoint coordinates and posture S2 are calculated. Similarly, when the right viewpoint calculation unit 203 receives the coordinates and orientation S1 of the virtual vehicle from the behavior identification unit 201, the right viewpoint calculation unit 203 calculates the installation position of the right virtual camera and the installation orientation of the yaw angle, roll angle, and pitch angle. , and the right viewpoint coordinates and posture S3 are calculated.

FIG. 5 is a configuration diagram of the left CG generation function and the right CG generation function.

The left CG generation function 3 has a left CG generation section 301 , a left CG transmission section 302 and CG environment information 7 . Similarly, the right CG generation function 4 has a right CG generation section 401 , a right CG transmission section 402 and CG environment information 7 .

The left CG generation unit 301 creates a CG space by CG based on the road information S12 input from the CG environment information 7 and the like. In this CG space, the left CG generator 301 creates a left CG 4 from the CG at the viewpoint specified by the left viewpoint coordinates and posture S 2 , and transmits the left CG 4 to the left CG transmitter 302 . The left CG transmission unit 302 adjusts the transmission timing of the left CGS4 based on the left CGS4 and the shutter signal S7 output from the ECU6.

The right CG generation unit 401 creates a computer CG space based on the road information S12 and the like input from the CG environment information 7 . In this CG space, the right CG generation unit 401 creates a right CGS5 from the CG at the viewpoint specified by the right viewpoint coordinates and posture S3, and transmits it to the right CG transmission unit 402 . The right CG transmission unit 402 adjusts the transmission timing of the right CGS5 based on the right CGS5 and the shutter signal S7 output from the ECU6.

FIG. 6 is a flow chart of the left CG generation function and the right CG generation function.

First, the left virtual camera will be explained. The left viewpoint calculation unit 202 calculates left viewpoint coordinates and attitude S2 (S602) using a shutter signal S7 transmitted by the ECU 6 at each shutter timing (S601) as a trigger. The left CG generator 301 creates the left CGS4 of the requested left viewpoint coordinates and orientation based on the calculated left viewpoint coordinates and orientation S2 (S603). The left CG generation section 301 transmits the created left CGS4 to the left CG transmission section 302 . The left CG transmission unit 302 receives left CGS4 (S604). The left CG transmission unit 302 does not output the left CGS4 to the CG conversion function 5 at the stage of receiving the left CGS4 but retains it. The left CG transmission unit 302 outputs the left CGS4 to the CG conversion function 5 at the timing of receiving the shutter signal S7 at the next shutter timing (S608) (S609).

Next, the virtual camera on the right will be described. The right CG generation unit 401 creates a right CGS5 based on the right viewpoint coordinates and the orientation S3 (S605) calculated by the right viewpoint calculation unit 203 for each shutter signal S7 at each shutter timing (S601) by the ECU 6 ( S606). The right CG generation unit 401 transmits the created right CGS5 to the right CG transmission unit 402 .
The right CG transmission unit 402 receives the right CGS5 (S607). The right CG transmission unit 402 does not output the right CGS5 by CG to the CG conversion function 5 at the stage of receiving the right CGS5, but holds it. The right CG transmission unit 402 outputs the right CGS5 to the CG conversion function 5 at the timing of receiving the shutter signal S7 at the next shutter timing (S608) (S610).

Thus, based on the shutter signal S7, the output timing of the left CGS4 and right CGS5 can be controlled, and the transmission timing of the left and right CGS4 and S5 to the CG conversion function 5 can be controlled.

FIG. 7 is a configuration diagram of the CG conversion function.

The CG conversion function 5 has a camera format conversion section 501 .

The camera format conversion unit 501 converts the left CGS4 input from the left CG transmission unit 302 and the right CGS5 input from the right CG transmission unit 402 into the camera output data format S6.

FIG. 8 is a configuration diagram of an ECU.

The ECU 6 has an image recognition section 601 and a vehicle control section 602 .

The image recognition unit 601 processes the left CGS4 and right CGS5 based on the camera output data format S6, and outputs the recognition result S13 such as the distance and speed of the object. Based on the recognition result S13 and the control information S11, the vehicle control unit 602 transmits information such as the accelerator signal S9, the brake signal S9, and the steering wheel operation signal S10, which are control feedback information for vehicle control, to the vehicle behavior simulator 1. output to

This configuration includes a vehicle behavior simulator 1 , a camera viewpoint specifying function 2 , a left CG generation function 3 and a right CG generation function 4 , and a CG conversion function 5 . The vehicle behavior simulator 1 calculates behavior information of a virtual vehicle. The camera viewpoint specifying function 2 calculates the viewpoint information of the virtual camera based on the CG request signal output from the ECU 6 and the behavior information of the virtual vehicle. A left CG generation function 3 and a right CG generation function 4 generate left CG and right CG based on a CG request signal output from the ECU 6 and viewpoint information of the virtual camera. A CG conversion function 5 converts the generated left CG and right CG into a format that can be input to the ECU 6 . As a result, by inputting the left CG and the right CG to the ECU 6 at the timing requested by the ECU 6 and without any time lag between the left and right sides, a parallax can be generated between the left and right virtual cameras, and the operation of the ECU 6 can be appropriately tested. can do.

The present invention is not limited to the above-described embodiments, and includes various modifications.

For example, the above-described embodiments have been described in detail in order to explain the present invention in an easy-to-understand manner, and are not necessarily limited to those having all the described configurations.

For example, in the HILS of the above-described embodiment, the stereo camera is used as the viewpoint of the virtual camera. The HILS is not limited to this, and may use one vehicle-mounted camera as the viewpoint of the virtual camera. In this case, the left CG generation function 3 and the left CG or the right CG generation function 4 and the right CG may be either the left CG generation function 3 and the left CG or the right CG generation function 4 and the right CG. As a result, the operation of the ECU can be appropriately tested even when a CG is input with a monocular vehicle-mounted camera as a viewpoint.

Furthermore, for example, the HILS may control a stereo camera that captures an image of the front of the vehicle and the ECU 6 of each of the other cameras. As a result, even if CG is input with each vehicle-mounted camera as a viewpoint, the operation of the ECU can be appropriately tested. In this case, the CG imaged by the stereo camera and the other camera may be input to the ECU 6 in synchronization.
Another camera may be a surrounding camera capable of imaging the surroundings of the vehicle.

1: vehicle behavior simulator, 2: camera viewpoint specifying function, 3: left CG generation function, 4: right CG generation function, 5: CG conversion function, 6: ECU, S1: vehicle coordinates and attitude, S2: left viewpoint coordinates and attitude, S3: right viewpoint coordinates and attitude, S4: left CG, S5: right CG, S7: shutter signal

Claims (9)

An electronic control unit that tests the operation of the electronic control unit by inputting CG (computer graphics) from a viewpoint of a virtual camera mounted on a virtual vehicle running on a virtual road built on a computer into the electronic control unit. An evaluation device,
a vehicle behavior simulator that calculates behavior information of the virtual vehicle;
a behavior identifying unit that identifies the behavior information used for generating the CG based on the CG request signal output from the electronic control unit , among the calculated behavior information ;
a viewpoint calculation unit that calculates viewpoint information of the virtual camera based on the identified behavior information of the virtual vehicle ;
a CG generation unit that generates the CG with the virtual camera as a viewpoint based on the viewpoint information of the virtual camera;
a CG transmission unit that transmits the generated CG in response to the CG request signal output from the electronic control unit;
and a CG conversion section for converting the CG transmitted from the CG transmission section into a format that can be input to the electronic control unit. comprising a plurality of the CG generation units and the CG transmission units ;
The virtual vehicle is equipped with a plurality of the virtual cameras,
The viewpoint calculation unit calculates viewpoint information of each virtual camera,
Each CG generation unit generates CG with each virtual camera as a viewpoint based on the viewpoint information of each virtual camera ,
2. The electronic control unit evaluation apparatus according to claim 1 , wherein each CG transmission section transmits the generated CG in response to the CG request signal output from the electronic control unit. 3. The evaluation apparatus for an electronic control unit according to claim 2, wherein said plurality of virtual cameras are stereo cameras that capture images in front of said virtual vehicle. 3. The evaluation apparatus for an electronic control unit according to claim 2, wherein said plurality of virtual cameras are a stereo camera that captures an image in front of said virtual vehicle and another camera. 5. The apparatus for evaluating an electronic control unit according to claim 4, wherein the CG from the viewpoints of the stereo camera and the another camera are inputted in synchronization with the electronic control unit. 2. The evaluation apparatus for an electronic control unit according to claim 1, wherein the viewpoint information of said virtual camera includes a mounting angle and attitude of said virtual camera with respect to said virtual vehicle. 2. The electronic control unit evaluation apparatus according to claim 1, wherein the behavior information of said virtual vehicle includes coordinates and attitude of said virtual vehicle in a virtual space. 2. The electronic control unit evaluation apparatus according to claim 1, wherein said CG request signal output from said electronic control unit is a shutter signal of a camera to be controlled by said electronic control unit. An electronic control unit that tests the operation of the electronic control unit by inputting CG (computer graphics) from a viewpoint of a virtual camera mounted on a virtual vehicle running on a virtual road built on a computer into the electronic control unit. An evaluation method,
calculating behavior information of the virtual vehicle;
identifying the behavior information to be used for generating the CG based on the CG request signal output from the electronic control unit , among the calculated behavior information;
calculating viewpoint information of the virtual camera based on the identified behavior information of the virtual vehicle;
generating the CG with the virtual camera as a viewpoint based on the viewpoint information of the virtual camera;
transmitting the generated CG in response to the CG request signal output from the electronic control unit;
An evaluation method for an electronic control unit, wherein the transmitted CG is converted into a format that can be input to the electronic control unit.

Images