JP2021072496A - Control method of stereo camera - Google Patents

Abstract

To provide a control method of a stereo camera which can reduce the size of a stereo camera.SOLUTION: In a control method of a stereo camera that includes first and second cameras that capture an object, and a memory that stores first image data acquired by the imaging of the first camera and second image data acquired by the imaging of the second camera, when a time from an imaging start time Ts1 of the first camera to a transfer end time Te1 of the first image data to the memory is T1, an imaging time of the second camera is Tex2, and T1-Tex2=ΔT1, the imaging start time Ts2 of the second camera is delayed by the ΔT1 with respect to the imaging start time Ts1.SELECTED DRAWING: Figure 2

Description

The present invention relates to a method for controlling a stereo camera.

In Patent Document 1, as a control method of a stereo camera having a first camera and a second camera, a synchronization signal is sent to each of the first camera and the second camera, and the timing of exposure start of the first camera and the second camera is determined. A method of synchronizingly imaging an object and transferring the captured image data to a memory is described.

JP-A-2019-129482

In the control method of Patent Document 1, since the timing of the exposure start of the first and second cameras is synchronized, the image data captured by the first and second cameras is simultaneously transferred to the memory. However, in order to simultaneously transfer the image data captured by the first and second cameras to the memory, it is necessary to increase the number of signal wirings. Therefore, in the configuration of Patent Document 1, the size of the stereo camera is increased by that amount.

The control method of the stereo camera of the present invention includes a first camera and a second camera that capture an object.
A method for controlling a stereo camera, which comprises a memory for storing a first image data acquired by imaging of the first camera and a second image data acquired by imaging of the second camera.
Let T1 be the time from the imaging start time Ts1 of the first camera to the transfer end time Te1 of the first image data to the memory.
Let Tex2 be the imaging time of the second camera.
When T1-Tex2 = ΔT1
It is characterized in that the imaging start time Ts2 of the second camera is delayed by the ΔT1 with respect to the imaging start time Ts1.

It is a figure which shows the whole structure of the robot system which concerns on 1st Embodiment. It is a block diagram which shows the whole structure of a stereo camera. It is a flowchart which shows the control method of a stereo camera. It is a flowchart which shows the control method of the stereo camera which concerns on 2nd Embodiment. It is a flowchart which shows the control method of the stereo camera which concerns on 3rd Embodiment.

Hereinafter, the control method of the stereo camera of the present invention will be described in detail based on the embodiments shown in the accompanying drawings.

FIG. 1 is a diagram showing an overall configuration of a robot system according to the first embodiment. FIG. 2 is a block diagram showing the overall configuration of the stereo camera. FIG. 3 is a flowchart showing a control method of the stereo camera.

The robot system 1 shown in FIG. 1 includes a robot 2, a stereo camera 4, a robot control device 5 that controls the drive of the robot 2 based on the measurement results of the stereo camera 4, and a host computer capable of communicating with the robot control device 5. 6 and. Each of these parts can be communicated by wire or wirelessly, and the communication may be performed via a network such as the Internet.

-Robot-
The robot 2 is, for example, a robot that performs operations such as supplying, removing, transporting, and assembling precision equipment and parts constituting the precision equipment. However, the use of the robot 2 is not particularly limited. The robot 2 of the present embodiment is a 6-axis robot, and as shown in FIG. 1, has a base 21 fixed to a floor or a ceiling, and a robot arm 22 connected to the base 21.

The robot arm 22 includes a first arm 221 rotatably connected to the base 21 around the first axis O1 and a second arm 222 rotatably connected to the first arm 221 around the second axis O2. , A third arm 223 rotatably connected to the second arm 222 around the third axis O3, and a fourth arm 224 rotatably connected to the third arm 223 around the fourth axis O4. It has a fifth arm 225 rotatably connected to the four arm 224 around the fifth axis O5, and a sixth arm 226 rotatably connected to the fifth arm 225 around the sixth axis O6. Further, the sixth arm 226 is equipped with an end effector 24 according to the work to be executed by the robot 2.

Further, the robot 2 has a first drive device 251 that rotates the first arm 221 with respect to the base 21, a second drive device 252 that rotates the second arm 222 with respect to the first arm 221 and a second. The third drive device 253 that rotates the third arm 223 with respect to the arm 222, the fourth drive device 254 that rotates the fourth arm 224 with respect to the third arm 223, and the fourth arm 224 with respect to the fourth arm 224. It has a fifth drive device 255 that rotates the fifth arm 225, and a sixth drive device 256 that rotates the sixth arm 226 with respect to the fifth arm 225. The first to sixth drive devices 251 to 256 each include, for example, a motor as a drive source, a controller for controlling the drive of the motor, and an encoder for detecting the amount of rotation of the motor. The first to sixth drive devices 251 to 256 are independently controlled by the robot control device 5, respectively.

The robot 2 is not limited to the configuration of the present embodiment, and the robot arm 22 may have, for example, one to five arms or seven or more arms. Further, for example, the type of the robot 2 may be a SCARA robot or a dual-arm robot having two robot arms 22.

-Robot control device-
The robot control device 5 receives a position command of the robot 2 from the host computer 6 and drives the first to sixth drive devices 251 to 256 so that the positions of the arms 221 to 226 correspond to the received position command. Each is controlled independently. The robot control device 5 includes, for example, a processor (CPU) composed of a computer and processing information, a memory communicatively connected to the processor, and an external interface. Various programs that can be executed by the processor are stored in the memory, and the processor can read and execute various programs and the like stored in the memory.

-Stereo camera-
As shown in FIG. 2, the stereo camera 4 is captured by the first camera 41 and the second camera 42 that capture the object X, and the first image data D1 and the second camera 42 that are captured by the first camera 41. A memory 43 for storing the second image data D2, an image processing control unit 44 for driving control of the first and second cameras 41 and 42, and image processing using the first and second image data D1 and D2. It has an output interface 45 that communicates with the host computer 6. In the present embodiment, the "image processing" means three-dimensional measurement of the object X.

Of these components, at least the first and second cameras 41 and 42 are fixed to the fifth arm 225 of the robot 2. Further, the first camera 41 and the second camera 42 are arranged at a predetermined distance from each other, and the relative positional relationship between them is fixed. Further, the first camera 41 and the second camera 42 are arranged so as to image the tip end side of the fifth arm 225, that is, the end effector 24 side, respectively.

Here, the relationship in which the end effector 24 is located on the tip end side of the fifth arm 225 is maintained even if the first to fourth arms 221 to 224 and the sixth arm 226 other than the fifth arm 225 move. Therefore, the stereo camera 4 can always take an image of the tip end side of the end effector 24. Therefore, the posture when the end effector 24 tries to grip the object X, that is, no matter what posture the end effector 24 faces the object X, the stereo camera 4 captures the object X in that posture. can do. Therefore, the three-dimensional measurement of the object X can be performed more reliably.

However, the arrangement of the stereo camera 4 is not particularly limited, and may be fixed to the first to fourth arms 221 to 224 and the sixth arm 226.

The first and second cameras 41 and 42 use, for example, a CMOS (Complementary Metal Oxide Semiconductor) sensor, a CCD (Charge Coupled Device) sensor, or the like. The driving methods of the first and second cameras 41 and 42 are not particularly limited, and may be any method such as a global shutter method that scans the entire area of the image at the same time, a rolling shutter method that sequentially scans the image for each scanning line, and the like. .. In the present embodiment, the first and second cameras 41 and 42 have the same configuration, but the configuration is not limited to this, and the first and second cameras 41 and 42 may have different configurations.

Further, the first camera 41 starts imaging of the object X based on the image pickup timing control signal Sd1 from the image processing control unit 44, and the second camera 42 starts the image pickup timing control signal Sd2 from the image processing control unit 44. The image of the object X is started based on the above. Then, the first image data D1 captured by the first camera 41 and the second image data D2 captured by the second camera 42 are immediately transferred to the memory 43 via the image processing control unit 44 after imaging, respectively. And saved. The above-mentioned "immediately after imaging" means that there is no delay after imaging, and for example, the data is transferred to the memory 43 before the first and second image data D1 and D2 obtained by the next imaging. , Means to be preserved.

The image processing control unit 44 reads the paired first and second image data D1 and D2 used for image processing from the memory 43, and performs stereo matching based on the parallax information of the read first and second image data D1 and D2. Three-dimensional measurement of the object X is performed. This measurement result is transferred to the memory 43 and stored.

Further, the image processing control unit 44 controls the imaging start times of the first and second cameras 41 and 42 so that the first and second image data D1 and D2 are not transferred to the memory 43 at the same time. In other words, the image processing control unit 44 has a time when the first image data D1 is transferred from the first camera 41 to the memory 43 and a time when the second image data D2 is transferred from the second camera 42 to the memory 43. The imaging start times of the first and second cameras 41 and 42 are delayed so that they do not overlap with each other. In particular, in the present embodiment, the image processing control unit 44 starts the transfer of the second image data D2 to the memory 43 after the transfer of the first image data D1 to the memory 43 is completed. , The imaging start time of the second cameras 41 and 42 is delayed.

According to such a configuration, the amount of data transferred at the same time can be reduced. More specifically, the amount of data transferred at the same time is the amount of image data for one image in the present invention, which was the amount of image data for two images in the past. Therefore, for example, the first and second image data D1 and D2 can be transferred to the memory 43 at a sufficiently high speed without expanding the bus width of the memory 43 or increasing the clock frequency. Therefore, it is possible to reduce the size and cost of the stereo camera 4 without delaying the image processing by the image processing control unit 44.

Hereinafter, the control method of the stereo camera 4 will be specifically described with reference to the flowchart shown in FIG. First, as step S11, the image processing control unit 44 acquires the time ΔT1 for delaying the imaging start time Ts2 of the second camera 42 with respect to the imaging start time Ts1 of the first camera 41. This delay time ΔT is calculated in advance and is stored in, for example, the memory 43. However, the storage location of the delay time ΔT1 is not particularly limited.

The delay time ΔT1 is calculated as follows. First, various settings of the first and second cameras 41 and 42, for example, ISO, aperture value, exposure time, and the like are set to conditions suitable for imaging the object X. Thereby, the imaging time Tex1 of the first camera 41 and the imaging time Tex2 of the second camera 42 are determined. The imaging time Tex1 can be said to be the exposure time of the first camera 41, and the imaging time Tex2 can be said to be the exposure time of the second camera 42. Further, in the present embodiment, since the first and second cameras 41 and 42 have the same configuration, the above-mentioned various settings are also the same, and in particular, Tex1 = Tex2. However, the present invention is not limited to this, and for example, Tex1 ≠ Tex2 may be used.

Next, the data obtained by actually driving the first camera 41 or performing a simulation to transfer the first image data D1 to the memory 43 from the imaging start time Ts1 when the imaging of the first camera 41 is started. The time T1 until the transfer end time Te1 is obtained. That is, T1 = Te1-Ts1. In this embodiment, the time required to transfer the first image data D1 to the memory 43 is constant. Then, the time ΔT1 to be delayed is obtained by subtracting the imaging time Tex2 of the second camera 42 from the calculated time T1. That is, ΔT1 = T1-Tex2. As described above, since Tex1 = Tex2 in this embodiment, ΔT1> 0.

Next, in step S12, the image processing control unit 44 transmits the image pickup timing control signal Sd1 to the first camera 41 to start the image pickup of the first camera 41. Next, in step S13, the image processing control unit 44 determines whether the time ΔT1 delayed from the imaging start time Ts1 has elapsed. Then, if the time ΔT1 delayed from the imaging start time Ts1 has not elapsed, step S13 is repeated. On the other hand, if the time ΔT1 delayed from the image pickup start time Ts1 has elapsed, as step S14, the image processing control unit 44 transmits the image pickup timing control signal Sd2 to the second camera 42 to capture the image of the second camera 42. Start. In this way, by delaying the imaging start time Ts2 of the second camera 42 with respect to the imaging start time Ts1 of the first camera 41 by the time ΔT1, the transfer of the first image data D1 from the first camera 41 to the memory 43 can be performed. After the completion, the transfer of the second image data D2 from the second camera 42 to the memory 43 is started. Therefore, as described above, it is possible to prevent the first and second image data D1 and D2 from being transferred to the memory 43 at the same time.

The time from the imaging start time Ts1 to the imaging start time Ts2 is preferably equal to the delay time ΔT1. As a result, the time from the imaging of the first camera 41 to the transfer of the second image data D2 to the memory 43 can be minimized. Therefore, the image processing by the image processing control unit 44 can be started more quickly. However, the time is not limited to this, and the time from the imaging start time Ts1 to the imaging start time Ts2 may be larger than the delay time ΔT1.

Further, in the present embodiment, since the delay time ΔT1> 0, the imaging start time Ts2 of the second camera 42 is later than the imaging start time Ts1 of the first camera 41, but is not limited to this, and ΔT1 < It may be 0 or ΔT1 = 0. If ΔT1 <0, the imaging start time Ts2 of the second camera 42 is earlier than the imaging start time Ts1 of the first camera 41, and if ΔT = 0, the imaging start time Ts2 of the second camera 42 is the first. The imaging start time of the camera 41 is the same as Ts1.

According to such a control method of the stereo camera 4, the imaging start time Ts1 and the imaging start time Ts2 deviate from each other. That is, the timings for acquiring the paired first and second image data D1 and D2 used for the three-dimensional measurement of the object X are different. Therefore, in the control method of the stereo camera 4, it is preferable that the object X is relatively stationary with respect to the stereo camera 4. If the object X is stationary, even if the timings for acquiring the first and second image data D1 and D2 are different, there is no deviation other than parallax between them, so that the object X can be accurately measured. Three-dimensional measurement can be performed. However, the present invention is not limited to this, and the object X may move relative to the stereo camera 4.

The control method of the stereo camera 4 has been described above. As described above, such a control method of the stereo camera 4 includes the first camera 41 and the second camera 42 that image the object X, and the first image data D1 and the second that are acquired by the imaging of the first camera 41. This is a control method for a stereo camera 4 including a memory 43 for storing the second image data D2 acquired by imaging the camera 42. In such a control method, the time from the imaging start time Ts1 of the first camera 41 to the transfer end time Te1 of the first image data D1 to the memory 43 is set to T1, and the imaging time of the second camera 42 is set to Tex2. When T1-Tex2 = delay time ΔT1, the time ΔT1 for delaying the imaging start time Ts2 of the second camera 42 with respect to the imaging start time Ts1 is delayed. With such a configuration, the first and second image data D1 and D2 are not transferred to the memory 43 at the same time, and the amount of data transferred to the memory 43 at the same time can be reduced. More specifically, the amount of data transferred at the same time is the amount of image data for one image in the present invention, which was the amount of image data for two images in the past. Therefore, for example, the first and second image data D1 and D2 can be transferred to the memory 43 at a sufficiently high speed without expanding the bus width of the memory 43 or increasing the clock frequency. Therefore, it is possible to reduce the size and cost of the stereo camera 4 without delaying the image processing by the image processing control unit 44.

Further, as described above, the object X is stationary relative to the stereo camera 4. As a result, even if the timings for acquiring the first and second image data D1 and D2 are different, there is no deviation other than parallax between them, so that the three-dimensional measurement of the object X can be performed accurately. it can.

<Second Embodiment>
FIG. 4 is a flowchart showing a control method of the stereo camera according to the second embodiment.

The present embodiment is the same as the first embodiment described above, except that the control method of the stereo camera 4 is different. In the following description, the present embodiment will be mainly described with respect to the differences from the first embodiment described above, and the description of the same matters will be omitted. Further, in FIG. 4, the same reference numerals are given to the same configurations as those of the above-described first embodiment.

In the control method of the stereo camera 4 of the first embodiment described above, the delay time ΔT1 is predetermined, whereas in the control method of the stereo camera 4 of the present embodiment, the first and second image data D1 and Before acquiring D2, preferably immediately before acquisition, the imaging environment of the first and second cameras 41 and 42 is detected, and the delay time ΔT1 is determined based on the detected imaging environment. The imaging environment means, for example, the brightness of the environment in which the object X is placed. As a result, when three-dimensional measurement of the object X is performed in an environment where the imaging environment changes over time, the environment is closer to the environment for acquiring the first and second image data D1 and D2. The delay time ΔT1 can be determined. Therefore, a more suitable delay time ΔT1 can be calculated.

Hereinafter, the control method of the stereo camera 4 will be specifically described with reference to the flowchart shown in FIG. First, in step S21, the image processing control unit 44 detects the imaging environment of the first and second cameras 41 and 42. The method for detecting the imaging environment is not particularly limited, but in the present embodiment, the first and second cameras 41 and 42 perform imaging for environment detection, and the imaging environment, that is, brightness is determined based on the obtained images. To detect. The detected brightness information is stored in the memory 43. According to such a method, the configuration of the stereo camera 4 becomes simple because other sensors and the like are not required. Further, since the imaging environment is detected based on the images captured by the first and second cameras 41 and 42 themselves, the imaging environment can be detected more accurately. However, the present invention is not limited to this, and for example, a method of detecting the imaging environment using a brightness sensor or a brightness sensor may be used.

When the first and second cameras 41 and 42 detect the imaging environment, it is preferable to perform the same control as the delay time ΔT1 as described above. That is, the first and second cameras so that the brightness information acquired by the imaging of the first camera 41 and the brightness information acquired by the imaging of the second camera 42 are not transferred to the memory 43 at the same time. It is preferable to delay the imaging start time of 41 and 42. In such a control method, the time from the imaging start time Tsf1 of the first camera 41 to the transfer end time Tef1 of the first image data D1 to the memory 43 is T1, the imaging time of the second camera 42 is Texf2, and Tf1. When −Texf2 = delay time ΔTf1, the time ΔTf1 for delaying the imaging start time Tsf2 of the second camera 42 with respect to the imaging start time Tsf1 is delayed. With such a configuration, the first and second image data D1 and D2 are not transferred to the memory 43 at the same time, and the amount of data transferred to the memory 43 at the same time can be reduced. However, it is not limited to this. Further, the imaging for detecting the imaging environment may be performed by any one of the first and second cameras 41 and 42.

Next, in step S22, the image processing control unit 44 calculates the delay time ΔT1. Specifically, the image processing control unit 44 first sets various settings of the first and second cameras 41 and 42, for example, ISO, aperture value, exposure time, etc., based on the imaging environment obtained in step S21. Set the conditions suitable for imaging the object X. As a result, the imaging times Tex1 and Tex2 of the first and second cameras 41 and 42 are determined. On the other hand, the time required to transfer the first image data D1 to the memory 43 does not fluctuate depending on the imaging environment and is constant, and is stored in the memory 43 in advance. Therefore, the image processing control unit 44 calculates the time T1 from the imaging time Tex1 of the first camera 41 and the time required to transfer the first image data D1 to the memory 43, and further sets the time T1 and the imaging time Tex2. The time to delay ΔT1 (= T1-Tex2) is calculated from.

Next, as step S23, the image processing control unit 44 transmits the image pickup timing control signal Sd1 to the first camera 41 to start the image pickup of the first camera 41. Next, in step S24, the image processing control unit 44 determines whether the time ΔT1 delayed from the imaging start time Ts1 has elapsed. If the time ΔT1 delayed from the imaging start time Ts1 has not elapsed, step S24 is repeated. On the other hand, if the time ΔT1 delayed from the imaging start time Ts1 has elapsed, as step S25, the image processing control unit 44 transmits the imaging timing control signal Sd2 to the second camera 42 to start imaging of the second camera 42. To do. In this way, by delaying the imaging start time Ts2 of the second camera 42 with respect to the imaging start time Ts1 of the first camera 41 by the time ΔT1, the transfer of the first image data D1 from the first camera 41 to the memory 43 can be performed. After the completion, the transfer of the second image data D2 from the second camera 42 to the memory 43 is started. Therefore, it is prevented that the first and second image data D1 and D2 are simultaneously transferred to the memory 43.

As described above, the control method of the stereo camera 4 of the present embodiment controls the delay time ΔT1 based on the imaging environment of the stereo camera 4, that is, the imaging environment of the first and second cameras 41 and 42. According to such a control method, the environment for acquiring the first and second image data D1 and D2 when three-dimensional measurement of the object X is performed in an environment where the imaging environment changes with time. The delay time ΔT1 can be determined based on a closer environment. Therefore, a more suitable delay time ΔT1 can be calculated.

Further, as described above, in the control method of the stereo camera 4 of the present embodiment, the brightness is detected as the imaging environment. Thereby, the imaging times Tex1 and Tex2 of the first and second cameras 41 and 42 suitable for the three-dimensional measurement of the object X can be set.

Even with such a second embodiment, the same effect as that of the first embodiment described above can be exhibited.

<Third Embodiment>
FIG. 5 is a flowchart showing a control method of the stereo camera according to the third embodiment.

The present embodiment is the same as the first embodiment described above, except that the control method of the stereo camera 4 is different. In the following description, the present embodiment will be mainly described with respect to the differences from the first embodiment described above, and the description of the same matters will be omitted. Further, in FIG. 5, the same reference numerals are given to the same configurations as those of the above-described first embodiment.

In the control method of the stereo camera 4 of the first embodiment described above, the image pickup start time Ts2 is delayed with respect to the image pickup start time Ts1 so that the first and second image data D1 and D2 are not transferred to the memory 43 at the same time. However, the time for the image processing control unit 44 to access the memory 43 and perform image processing based on the first and second image data D1 and D2 is not considered. Therefore, the time during which the image processing control unit 44 performs image processing overlaps with the time during which the first and second image data D1 and D2 are transferred to the memory 43, which may reduce the transfer speed and processing speed. There is. Therefore, in the present embodiment, the imaging start times Ts1 and Ts2 are determined in consideration of the time during which the image processing control unit 44 is performing image processing so that they do not overlap.

Hereinafter, the control method of the stereo camera 4 will be specifically described with reference to the flowchart shown in FIG. First, in step S31, the image processing control unit 44 acquires the delay times ΔT1 and ΔT2. The delay times ΔT1 and ΔT2 are calculated in advance and are stored in, for example, the memory 43. However, the storage location of the delay times ΔT1 and ΔT2 is not particularly limited.

The method for calculating the delay time ΔT1 is the same as that in the first embodiment described above. On the other hand, the delay time ΔT2 is calculated as follows. First, various settings of the first and second cameras 41 and 42, for example, ISO, aperture value, exposure time, and the like are set to conditions suitable for imaging the object X. Thereby, the imaging time Tex1 of the first camera 41 is determined. Further, the time T3 required for the image processing of the image processing control unit 44 is calculated by actually driving the image or performing a simulation. The time T3 is the time from the start time Ts3 when the image processing control unit 44 accesses the memory 43 and starts the image processing to the end time Te3 when the result of the image processing is stored in the memory 43 and the image processing is finished. is there. That is, T3 = Te3-Ts3. Then, the delay time ΔT2 is obtained by subtracting the imaging time Tex1 of the first camera 41 from the calculated time T3. That is, ΔT2 = T3-Tex1.

Next, as step S32, the image processing control unit 44 accesses the memory 43 and starts image processing. Next, in step S33, the image processing control unit 44 determines whether or not the time ΔT2 to be delayed from the start time Ts3 has elapsed. If the time ΔT2 to be delayed from the start time Ts3 has not elapsed, step S33 is repeated. On the other hand, if the time ΔT2 delayed from the start time Ts3 has elapsed, as step S34, the image processing control unit 44 transmits the image pickup timing control signal Sd1 to the first camera 41 to start imaging of the first camera 41. To do. By delaying the imaging start time Ts1 with respect to the start time Ts3 by ΔT2 in this way, the transfer of the first image data D1 from the first camera 41 to the memory 43 is started after the image processing is completed.

Next, in step S35, the image processing control unit 44 determines whether the time ΔT1 delayed from the imaging start time Ts1 has elapsed. If the time ΔT1 delayed from the imaging start time Ts1 has not elapsed, step S35 is repeated. On the other hand, if the time ΔT1 delayed from the image pickup start time Ts1 has elapsed, as step S36, the image processing control unit 44 transmits the image pickup timing control signal Sd2 to the second camera 42 to capture the image of the second camera 42. Start. By delaying the imaging start time Ts2 with respect to the imaging start time Ts1 in this way, the memory from the second camera 42 is completed after the transfer of the first image data D1 from the first camera 41 to the memory 43 is completed. The transfer of the second image data D2 to 43 is started.

According to such a control method of the stereo camera 4, the transfer of the first image data D1 to the memory 43, the transfer of the second image data D2 to the memory 43, and the image processing by the image processing control unit 44 are time-consuming. It doesn't become. Therefore, it is possible to effectively suppress a decrease in transfer speed and processing speed. For example, the first is performed at a sufficiently high speed without increasing the bus width of the memory 43 or increasing the clock frequency. , The second image data D1 and D2 can be transferred to the memory 43, and the image processing by the image processing control unit 44 can be performed.

The time from the start time Ts3 to the imaging start time Ts1 is preferably equal to the delay time ΔT2. As a result, the repetition period of image processing can be minimized. However, the time is not limited to this, and the time from the start time Ts3 to the imaging start time Ts1 may be larger than the delay time ΔT2.

As described above, the stereo camera 4 has an image processing control unit 44 that performs image processing using the first image data D1 and the second image data D2 stored in the memory 43. Then, in the control method of the stereo camera 4 of the present embodiment, the time from the start time Ts3 of the image processing to the end time Te3 of the image processing is set to T3, the imaging time of the first camera 41 is set to Tex1, and T3-Tex1 = delay. When the time ΔT2 is set, the time ΔT2 for delaying the imaging start time Ts1 with respect to the start time Ts3 is delayed. According to such a configuration, the transfer of the first image data D1 to the memory 43, the transfer of the second image data D2 to the memory 43, and the image processing by the image processing control unit 44 do not overlap in time. Therefore, it is possible to effectively suppress a decrease in transfer speed and processing speed. For example, the first is performed at a sufficiently high speed without increasing the bus width of the memory 43 or increasing the clock frequency. , The second image data D1 and D2 can be transferred to the memory 43, and the image processing by the image processing control unit 44 can be performed.

Even with such a third embodiment, the same effect as that of the first embodiment described above can be exhibited.

Although the control method of the stereo camera of the present invention has been described above based on the illustrated embodiment, the present invention is not limited to this, and the configuration of each part is an arbitrary configuration having the same function. Can be replaced with. Further, any other constituents may be added to the present invention.

In the above-described embodiment, the configuration in which the stereo camera 4 is mounted on the robot 2 has been described, but the present invention is not limited to this, and for example, the stereo camera 4 is mounted on an automobile and used for driving assistance, automatic driving, and the like. You may.

1 ... Robot system, 2 ... Robot, 21 ... Base, 22 ... Robot arm, 221 ... 1st arm, 222 ... 2nd arm, 223 ... 3rd arm, 224 ... 4th arm, 225 ... 5th arm, 226 ... 6th arm, 24 ... End effector, 251 ... 1st drive device, 252 ... 2nd drive device, 253 ... 3rd drive device, 254 ... 4th drive device, 255 ... 5th drive device, 256 ... 6th drive device 4, 4 ... Stereo camera, 41 ... 1st camera, 42 ... 2nd camera, 43 ... Memory, 44 ... Image processing control unit, 45 ... Output interface, 5 ... Robot control device, 6 ... Host computer, D1 ... 1st image Data, D2 ... 2nd image data, O1 ... 1st axis, O2 ... 2nd axis, O3 ... 3rd axis, O4 ... 4th axis, O5 ... 5th axis, O6 ... 6th axis, S11-1 to S14, S21 ~ S25, S31 ~ S36 ... Step, Sd1, Sd2 ... Imaging timing control signal, X ... Object

Claims (6)

The first and second cameras that capture the object,
A method for controlling a stereo camera, which comprises a memory for storing a first image data acquired by imaging of the first camera and a second image data acquired by imaging of the second camera.
Let T1 be the time from the imaging start time Ts1 of the first camera to the transfer end time Te1 of the first image data to the memory.
The imaging time of the second camera is set to Tex2.
When T1-Tex2 = ΔT1
A method for controlling a stereo camera, which comprises delaying the imaging start time Ts2 of the second camera with respect to the imaging start time Ts1 by ΔT1. The method for controlling a stereo camera according to claim 1, wherein the ΔT1 is controlled based on the imaging environment of the stereo camera. The method for controlling a stereo camera according to claim 2, wherein the brightness is detected as the imaging environment. In the detection of the imaging environment,
The time from the imaging start time Tsf1 of the first camera to the transfer end time Tef1 of the first image data to the memory is defined as Tf1.
The imaging time of the second camera is Texf2.
When Tf1-Texf2 = ΔTf1
The stereo camera control method according to claim 3, wherein the imaging start time Tsf2 of the second camera is delayed by ΔTf1 with respect to the imaging start time Tsf1. The method for controlling a stereo camera according to any one of claims 1 to 4, wherein the object is stationary relative to the stereo camera. Let T3 be the time from the start time Ts3 of the image processing using the first image data and the second image data to the end time Te3 of the image processing.
The imaging time of the first camera is set to Tex1.
When T3-Tex1 = ΔT2
The method for controlling a stereo camera according to any one of claims 1 to 5, wherein the imaging start time Ts1 is delayed with respect to the start time Ts3 by ΔT2.

Images