JP3928871B2 - Identification apparatus and method - Google Patents

Description

The present invention relates to the identification device and Methods, for example, it is suitably applied to the individual identification system.

In recent years, as one of the mobile robots, research on so-called autonomous mobile robots that have been able to take in information about the surrounding environment sequentially and decide their own actions based on the obtained information and move. (For example, refer to Patent Document 1).
JP 2001-157978 A

  By the way, as a method for such a mobile robot to identify an individual of another mobile robot (for example, No. 1, No. 2, No. 2, etc.), each robot is individually identified for radio waves, infrared rays, sound waves, etc. A first method for generating a special signal, a second method for applying a predetermined color different to each robot, and identifying the robot based on the color, and a symbol or bar for identification on the surface of each robot A third method for expressing a code or the like is conceivable.

  However, in the first method, a special device for transmitting or receiving a signal is required, and a desired signal may be transmitted due to restrictions by radio wave regulations or influence on other devices existing in the vicinity. There is a problem that may not be possible.

  In the second method, if only a small number of robots are to be identified, there is no problem because it is sufficient to prepare colors that are easy to identify as many as the number of robots. Differences must be used, and image processing becomes complicated, and there is a problem that it is easily affected by lighting conditions and the like.

  Furthermore, the third method has a problem that symbols or the like cannot be observed depending on the orientation or posture of the robot.

The present invention has been made in view of the above, it is intended to propose a simple identification device and Methods can reliably identify the mobile.

  In order to solve such a problem, in the present invention, in the identification device, identification bodies having different color patterns provided on each moving body, and movements provided on each moving body and disposed on other moving bodies, respectively. An image pickup means for picking up an identification object having a different color pattern for each body, a color pattern detection means for detecting a color pattern of the picked-up identification object based on first image information supplied from the image pickup means, and the detection Based on the result and the color pattern information of the identifying object for each moving object stored in advance, an identifying means for identifying the moving object having the identifying object imaged by the imaging means is provided.

  As a result, with this identification device, it is possible to easily and reliably identify each individual moving object.

  Further, in the present invention, in the identification device, an imaging unit that is provided in each moving body and captures an identification body having a different color pattern for each moving body disposed in the other moving body, and the imaging unit supplies the imaging unit. Based on the color pattern detection means for detecting the color pattern of the captured identification object based on the first image information, the detection result, and the color pattern information of the identification object for each moving object stored in advance. Thus, an identification means for identifying a moving body having an identification body imaged by the imaging means is provided.

  As a result, with this identification device, it is possible to easily and reliably identify each individual moving object.

  Furthermore, in the present invention, in the identification method, the first step of providing each moving body with an identification body having a different color pattern, and the imaging means provided for each moving body, the other moving body is identified. A second step of imaging, a third step of detecting a color pattern of the imaged identification object based on the first image information output from the imaging means, a color pattern of the detected identification object, and pre-stored And a fourth step of identifying the imaged identification object based on the color pattern information of each identification object.

  As a result, according to this identification method, it is possible to easily and reliably identify each moving object.

  Further, according to the present invention, in the identification device, for each moving body respectively given to a predetermined position of each moving body based on the imaging means for imaging the entire area and the first image information output from the imaging means. Color pattern detection means for detecting different color patterns, and identification means for identifying each moving body based on the detection result and information on the color pattern assigned to each moving body stored in advance are provided. I did it.

  As a result, with this identification device, it is possible to easily and reliably identify each individual moving object.

  Further, according to the present invention, in the identification method, a first step of arranging an imaging unit for imaging the entire region at a predetermined position and giving a different color pattern to each predetermined position of each moving body, and an output from the imaging unit A second step of detecting the color pattern of each mobile object based on the first image information, the detection result, and information on the color pattern assigned to each mobile object stored in advance, And a third step for identifying each moving object.

  As a result, according to this identification method, it is possible to easily and reliably identify each moving object.

According to the present invention as described above, Oite the identification device and Methods for identifying a movable body that moves a predetermined area, each mobile is provided with identification of the different color patterns from each other, each mobile The image pickup means provided in each image picks up an image of an identification object of another moving body , detects a color pattern of the identification object picked up by the image pickup means based on the first image information output from the image pickup means, A simple identification device capable of reliably identifying a moving body by identifying the moving body imaged by the imaging means based on the detection result and the color pattern information of each identifying body stored in advance. the Oyobi how method can be realized.

  Further, in an identification apparatus and method for identifying a plurality of moving objects that move within a predetermined area, an imaging unit that images the entire area is arranged at a predetermined position, and a different color pattern is given to each predetermined position of each moving object. Then, based on the first image information output from the imaging means, the color pattern of each moving body is detected, and the detection result and information on the color pattern provided for each moving body stored in advance are detected. By identifying each moving body based on the above, it is possible to realize a simple identification device and method that can reliably identify the moving body.

  Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings.

(1) First embodiment (1-1) Overall configuration of an individual identification system according to the first embodiment In FIG. 1, reference numeral 1 denotes an individual identification system to which the present invention is applied as a whole. Robots 3A to 3C are arranged. In the following description, it is assumed that the region 2 has a flat rectangular shape, a direction parallel to a predetermined side is an X direction (arrow x), a direction perpendicular thereto is a Y direction (arrow y), and the region 2 The vertical direction is defined as the Z direction (arrow z).

  In each of the robots 3 </ b> A to 3 </ b> C, as shown in FIG. 2, a head 12 is disposed at the front end of the upper surface of the body portion 10 via the neck portion 11, and at the front, rear, left and right corners of the lower surface of the body portion 10. A right front leg portion 15A, a left front leg portion 15B, a right rear leg portion 15C, and a left rear leg portion 15D (hereinafter collectively referred to as the leg portions 15A to 15D) each including a crotch portion 13 and a shin portion 14 are disposed. Is configured.

  In this case, a camera 16 is attached to the head 12, a microphone 17 (FIG. 3) is attached to a predetermined position of the head 12, and the head 12, the body part 10, and the legs 15 </ b> A to 15 </ b> D are attached. A plurality of contact sensors 18A to 18Z (FIG. 3) are respectively attached to the surface.

  Further, a control unit 19 (FIG. 3) is disposed in the body unit 10, and the control unit 19 is supplied from an image signal S1 supplied from the camera 16 and a microphone 17, as shown in FIG. The surrounding state is recognized based on the audio signal S2 and the sensor signals S3A to S3Z respectively supplied from the contact sensors 18A to 18Z, and based on the recognition result, each component unit (head 12, neck 11, trunk 10, The actuators 21A to 21J in the joint portions 20A to 20J (FIG. 2) connecting the respective crotch portions 13 and the shin portions 14) are driven as necessary.

  Thereby, in each robot 3A-3C, each component unit can be freely driven under the control of the control unit 19, so that an action corresponding to the surrounding environment can be taken autonomously. Has been made.

  In addition to such a configuration, in the case of this individual identification system 1, as is clear from FIG. 2A, each robot 3A is connected to the rear end portion of the upper surface of the body portion 10 of each robot 3A-3C via a support rod 22, respectively. A spherical discriminating body 23 that is color-coded with a different pattern every 3C is attached.

  As shown in FIG. 2 (B), each identifier 23 puts predetermined three colors on the surface of the sphere in a direction perpendicular to the moving direction of the robots 3A to 3C (that is, the Z direction). It is formed by applying the strips side by side.

  Further, in the control unit 19 (FIG. 3) of each robot 3A to 3C, the individual robots 3A to 3C moving in the region 2 are identified based on the identification body 23, as shown in FIG. An individual identification unit 30 is provided.

  That is, the individual identification unit 30 has the same number (for example, 16) of color extraction units 31A to 31U as the colors used for identification, and the image signal S1 supplied from the camera 16 is used as the color extraction units 31A to 31U. To enter each.

  Each color extraction unit 31 extracts a pixel of a predetermined color from the image based on the image signal S1, a portion corresponding to the pixel rises to a logical “1” level, and a portion corresponding to the other pixels is a logical “1”. The color extraction signals S10A to S10U that have fallen to the “0” level are sent to the color pattern detection unit 32, respectively. Each of the color extraction units 31A to 31U extracts pixels having different colors from a plurality of colors used for identification.

  The color pattern detection unit 32 superimposes and scans the images obtained based on the color extraction signals S10A to S10U supplied from the color extraction units 31A to 31U, so that the colors in the image are substantially circular in the Z direction. A portion in which three colors are arranged is detected, this is determined to be the identification body 23, and the color pattern of the identification body 23, the position in the image, and the diameter in units of pixels are used as the identification body information signal S11. The data is sent to the comparison / calculation unit 33.

  The comparison / calculation unit 33 calculates the color pattern of the imaged identification object 23 obtained based on the supplied identification object information signal S11, the color pattern of each identification object 23 stored in advance in the first memory 34, and the color pattern thereof. Based on the table relating to the ID, the ID of the identification body 23 captured by the camera 16 (that is, the individual of the robots 3A to 3C) is detected.

  Further, the comparison / calculation unit 33 stores the diameter (Dpic) in units of pixels in the image of the picked-up identifier 23 obtained based on the supplied identifier information signal S11 and the second memory 35 in advance. Based on the stored reference value (the diameter (Dstd) in units of pixels of the identification body 23 when the camera 16 and the identification body 23 are separated by 1 [m]), the following formula

The distance L1 to the identification body 23 is calculated by performing the above calculation.

  Further, the comparison / calculation unit 33 identifies the ID of the detected identifier 23, the distance L1 to the identifier 23 obtained by the calculation of the equation (1), and the identification in the image obtained based on the identifier information signal S11. The position of the body 23 is sent as an identification body detection signal S12 to the uppermost CPU 36 that controls the actions of the robots 3A to 3C as a whole.

  In this way, in the robots 3A to 3C, the CPU 36 can recognize the positions of the other robots 3A to 3C based on the identification object detection signal S12, and thus more appropriately according to the surrounding situation based on the recognition result. You can take action.

(1-2) Configuration of Color Extraction Unit Here, each color in an image based on an image signal is generally one of two color difference signals (RY, BY) included in the image signal. It can be expressed as a point on the UV plane where the signal level U of the first color difference signal is taken on the X axis and the signal level V of the other second color difference signal is taken on the Y axis. However, even on the same color, the position on the UV plane changes slightly depending on the illumination conditions.

  Therefore, as shown in FIG. 5, a rectangular area 40 in the UV plane is considered, and when the first and second color difference levels U and V of the pixel enter the area 40, the color that the pixel is to extract is displayed. By moving the area 40 according to the luminance level Y of the pixel in order to determine that it is present and to respond to a change in the illumination condition or the like, it is possible to accurately extract the pixel of the corresponding color in the image based on the image signal. .

In this case, in order to specify the area 40, the upper limit values a ₁ and b ₁ and the lower limit values a ₂ and b ₂ of the first and second color difference levels U and V may be determined according to the color to be extracted. In order to move the area 40 according to the luminance level Y, the upper limit values a ₁ and b ₁ and the lower limit values a ₂ and b ₂ of the first and second color difference levels U and V are calculated and determined for each luminance level Y in advance. Then, a table is created, and the upper limit values a ₁ and b ₁ and the lower limit values a ₂ and b ₂ of the first and second color difference levels U and V are changed based on the table and the luminance level Y of the actual pixel. You can make it.

  Considering the above points, the color extraction units 31A to 31U of the individual identification unit 30 of the robots 3A to 3C are configured as shown in FIG.

  That is, in the color extraction units 31A to 31U, the image signal S1 supplied from the camera 16 is input to the separation circuit 41, and the separation circuit 41 converts the image signal S1 into the luminance signal S20 and the two color difference signals (RY, B). -Y) Separate into S21A and S21B.

  In this case, the luminance signal S20 is sampled at a predetermined first period in the analog / digital conversion circuit 42 to be sequentially converted into luminance data D1 corresponding to each pixel, and thereafter the first to fourth memories. 44A to 44D.

  Further, one of the two color difference signals S21A and S21B is sampled at the above-described first period in the analog / digital conversion circuit 43A, whereby the first color difference signal corresponding to each pixel is obtained. The data D2A is sequentially converted and then supplied to the first and second comparison circuits 45A and 45B, and the other second color difference signal S21B is sampled in the analog / digital conversion circuit 43B at the above-described first period. As a result, the second color difference data D2B corresponding to each pixel is sequentially converted and then supplied to the third and fourth comparison circuits 45C and 45D.

At this time, the first and second memories 44A and 44B correspond to the colors to be extracted by the color extraction units 31A to 31U, and the first value for each value of the luminance data D1 (that is, the luminance level Y) of each pixel. with an upper limit value a ₁ or a lower limit value a ₂ are stored as a table of the color difference data D2A value (i.e. first chrominance level U), the third and fourth memory 44C, the 44D, the luminance of each pixel upper limit value b ₁ or the lower limit value b ₂ of the second color difference data D2B value for each value of the data D1 (that is, the second chrominance level V) is stored as a table. The upper limit values a ₁ and b ₁ and the lower limit values a ₂ and b ₂ are set by the uppermost CPU 36 described above via the address bus 46 and the data bus 47 using the luminance level Y of the pixel as an address.

  The first to fourth memories 44A to 44D output corresponding setting values from a table stored in advance with the value of the supplied luminance data D1 as an address.

Thus, in the color extraction units 31A to 31U, the upper limit value a ₁ or the lower limit of the first color difference level set in advance corresponding to the value of the luminance data D1 for each pixel from the first and second memories 44A to 44D. The value a ₂ is output, and the upper limit value b ₁ and the lower limit value of the second color difference level set in advance according to the value of the luminance data D1 for each pixel from the third and fourth memories 44C and 44D, respectively. b ₂ is output, they are supplied to the corresponding first to fourth comparison circuit 45A to 45D.

  In the first to fourth comparison circuits 45A to 45D, the outputs of the corresponding first to third or fourth memories 44A to 44D and the first or second color difference data D2A for each pixel sequentially supplied. , D2B are sequentially compared with each other, and the comparison results are sequentially sent to the determination circuit 48.

The determination circuit 48 has, for example, an AND circuit configuration, and based on the outputs of the first to fourth comparison circuits 45A to 45D, the pixels are stored in the first to fourth memories 44A to 44D, respectively. 1 or the second chrominance level U, determines whether or not contained in the area 40 to be determined (Fig. 5) by the upper limit value _a 1, _{b 1} or the lower limit value _a 2, _{b 2} and V, containing “1” is stored in the case, and “0” is stored in the frame memory 49 at a position corresponding to the pixel.

  Thus, the frame memory 49 outputs color extraction signals S10A to S10U in which only the portions corresponding to the color pixels to be extracted by the color extraction units 31A to 31U rise to the logic “1” level.

  In this way, each of the color extraction units 31A to 31U can extract the corresponding color pixel from the image based on the image signal S1, and the color extraction signals S10A to S10U thus obtained are color-coded as described above. The data is sent to the pattern detection unit 32 (FIG. 4).

(1-3) Operation and Effect of First Example In the above configuration, in the case of this individual identification system 1, each robot 3 </ b> A to 3 </ b> C uses the color pattern of the identification body 23 captured by the camera 16 and the camera 16. The position in the image based on the supplied image signal S1 and the diameter in units of pixels are detected by the color extraction units 31A to 31U and the color pattern detection unit 32 of the individual identification unit 30, and the detected identification object 23 is detected. And the IDs of the identification bodies 23 (individuals of the robots 3A to 3C) captured based on the color patterns of the identification bodies 23 and the color patterns of the identification bodies 23 stored in the first memory 34 and the table relating to the IDs, The distance to the identification object 23 based on the diameter in pixels of the identification object 23 detected by the color pattern detection unit 32 and the reference value stored in the second memory 35 L1 is calculated, and the highest CPU 36 that controls the behavior of the robots 3A to 3C, respectively, detects the ID of the identifier 23 detected, the distance L1 to the identifier 23, and the position of the identifier 23 in the image. To send.

  Therefore, in this individual identification system 1, the uppermost CPU of each robot 3A to 3C recognizes the individual of the other robots 3A to 3C located around each other and the distance L1 and direction to the robots 3A to 3C with high accuracy. can do.

  In addition, in this individual identification system 1, the identification body 23 of each of the robots 3A to 3C can be manufactured by painting a color such as a ping-pong ball, so that it can be constructed very inexpensively.

  Furthermore, in this individual identification system 1, since each robot 3A-3C does not use a special signal as a means for allowing the individual robots 3A-3C to identify individual robots, it adversely affects peripheral devices, There are no regulations based on regulations.

  Furthermore, in this individual identification system 1, since the identification body 23 is spherical as described above and a plurality of identification colors are applied in a band shape in a direction perpendicular to the moving direction of the robots 3A to 3C, the robots 3A to 3C. Can be seen in the same shape and the same color pattern from any direction, so that each robot 3A-3C is an individual of the other robots 3A-3C. Can be easily and reliably identified.

  Furthermore, in this individual identification system 1, the identification bodies 23 of the robots 3A to 3C are configured such that the surface is painted with a plurality of colors instead of a single color, so that a large number of robots 3A to 3C are identified according to how the colors are combined. It can be done. In this case, the number of combinations can be increased even if only colors with greatly different hues such as primary colors are used as identification colors, which affects lighting conditions, etc. compared to individual identification due to subtle color differences. There is an advantage that is difficult to receive.

  According to the above configuration, the identification bodies 23 having different color patterns are attached to the robots 3A to 3C, respectively, while the color patterns of the identification bodies 23 captured by the camera 16 in the robots 3A to 3C and the cameras 16 The position and the diameter in the image based on the image signal S1 supplied from the image are respectively detected, the detection result, the color pattern of each identifier 23 stored in advance in the first memory 34, and the table relating to the ID. Based on the diameter of the identification body 23 in the image based on the detected image signal S1 and the reference value stored in the second memory 35. By calculating the distance L1 to the discriminator 23, each robot 3A to 3C can easily and reliably identify the individual robots 3A to 3C. It can be identified, thus can realize individual identification system and a robot having a simple structure capable of recognizing reliably individual robots 3A-3C.

(2) Second Embodiment (2-1) Overall Configuration of Individual Identification System According to Second Embodiment FIGS. 7 and 8 show an individual identification system 50 according to the second embodiment, and FIG. As shown in FIG. 9 in which the same reference numerals are assigned to the corresponding parts, identification seals 52 are attached to the upper surfaces of the body parts 10 of the robots 51A to 51C.

  In this case, as shown in FIG. 10, the identification stickers 52 of the robots 51A to 51C are each divided into a predetermined number (for example, six) of belt-shaped regions 52A to 52F whose surfaces are sequentially adjacent to each other. 52F is configured by being colored in any one of a plurality of colors selected for identification. Further, the combination of colors (color patterns) for coloring the strip regions 52A to 52F on the surface of the identification seal 52 is selected for each identification seal 52. Thus, the robot 51A is based on the color pattern of the identification seal 52. ˜51C individuals can be identified.

  On the other hand, as is clear from FIGS. 7 and 8, a camera 53 is disposed above the area 2 so that the entire area 2 can be photographed on one screen, and an image output from the camera 53 is displayed. The signal S30 is supplied to the individual identification unit 54 disposed outside the region 2.

  In this case, the individual identification unit 54 extracts a number of colors (16 colors) corresponding to the number of colors for identification, as shown in FIG. Units 31A to 31U are provided, and color extraction signals S10A to S10U output from the color extraction units 31A to 31U are sent to the color pattern detection unit 55, respectively.

  The color pattern detection unit 55 detects a portion where a predetermined number of colors for identification are arranged in a strip shape in the image by overlapping and scanning the images based on the supplied color extraction signals S10A to S10U. The portion is determined as the identification sticker 52 of the robots 51A to 51C, and the color pattern of the identification sticker 52 and the position in the image in units of pixels are sent to the comparison / calculation unit 56 as an identification sticker detection signal S31. . In this case, the identification stickers 52 are detected by the number of the robots 51A to 51C located in the area 2, and the color pattern of each identification sticker 52 and the position in the image are sent to the comparison / calculation unit 56, respectively.

  The comparison / calculation unit 56 includes a color pattern of each identification sticker 52 obtained based on the supplied identification sticker detection signal S31, a color pattern of each identification sticker 52 stored in the memory 57 in advance, and a table relating to the ID. The ID of each identification sticker 52 in the image based on the image signal S30 from the camera 53 (that is, the individual of each robot 51A to 51C) is detected, and the ID of each identification sticker 52 thus obtained and its region 2 are detected. The position information inside is transmitted as a robot position detection signal S32 by radio waves to each of the robots 51A to 51C moving in the area 2 via the transmission unit 58.

  In each of the robots 51A to 51C, as shown in FIG. 12, in which parts corresponding to those in FIG. 3 are denoted by the same reference numerals, radio waves transmitted from the transmitting unit 58 of the individual identifying unit 54 are received by the receiving unit 61 via the antenna 60. The received IDs of the respective identification seals 52 (that is, the individual robots 51A to 51C) and the position information in the region 2 are sent to the control unit 62 as reception signals S33.

  At this time, the control unit 62 is supplied with the image signal 1 from the camera 16, the audio signal S2 from the microphone 17, and the sensor signals S3A to S3Z from the contact sensors 18A to 18Z.

  Thus, the control unit 62 determines the surrounding environment and the positions of the robots 51A to 51C including itself in the area 2 based on the supplied image signal S1, audio signal S2, sensor signals S3A to 3Z and reception signal S33. Can be recognized, the action is determined by itself based on the recognition result, and the corresponding actuators 21A to 21J are driven based on the determination result.

  In this way, each robot 51A to 51C recognizes the surrounding environment and the positions of all the robots 51A to 51C including itself moving in the region 2, and acts autonomously based on the recognition result. Has been made to get.

(2-2) Operation and effect of the second embodiment In the above configuration, in the individual identification system 50, the entire region 2 is imaged by the camera 53 from above, and the region 2 is based on the obtained image signal S30. The position of each identification seal 52 in the image based on the image signal S30 is detected by the individual color extraction units 31A to 31U and the color pattern detection unit 55 by the individual identification unit 54 arranged outside the image signal S30. The position and ID of each identification sticker 52 in the image is detected from the color pattern information of each identification sticker 52 stored therein, and the detection result is transmitted to each robot 51A to 51C in the region 2 via the transmission unit 58. To do.

  On the other hand, in each of the robots 51A to 51C, the position information in the region 2 of each robot 51A to 51C transmitted from the individual identification unit 54, the image signal S1 supplied from the camera 16, and the audio signal supplied from the microphone 17 Based on S2 and the sensor signals S3A to S3Z respectively supplied from the contact sensors 18A to 18Z, the surrounding situation and the positions of the robots 51A to 51C are recognized, and autonomously act based on the recognition result.

  Therefore, in the individual identification system 50, each robot 51A to 51C can surely recognize the other robots 51A to 51C in the region 2 and the absolute position of itself.

  In the individual identification system 50, the robots 51A to 51C are provided with the antenna 60 and the receiving unit 61 and only need to have the identification seal 52 attached thereto. This can be simplified as compared with the robots 3A to 3C (FIG. 2A) of the first embodiment. In addition, the system as a whole requires a camera 53 and an individual identification unit 54 for imaging the entire region 2, but this is sufficient regardless of the number of robots 51 </ b> A to 51 </ b> C. The configuration can be simplified.

  Further, in this individual identification system 50, the surface of the identification seal 52 is painted with a plurality of colors instead of a single color, so that a large number of robots 51A to 51C can be easily identified depending on how the colors are combined. . In this case, the number of combinations can be increased even if a color with a greatly different hue such as a primary color is used as a color for identification, compared to the case where individual identification is performed from a subtle color difference using a single color. Therefore, there is an advantage that it is not easily affected by lighting conditions.

  Further, in this individual identification system 50, since the ID and position of each robot 51A to 51C in the region 2 can be intensively recognized by the individual identification unit 54, each robot 51A to 51C is stored by storing this ID. Can be used for evaluation and improvement of programs for controlling the robots 51A to 51C.

  Furthermore, since this individual identification system 50 does not use the vision of each of the robots 51A to 51C, even when each of the robots 51A to 51C has a low visual processing capability or when each of the robots 51A to 51C does not have any vision. There are also advantages that can be applied.

  According to the above configuration, the entire action region 2 of the robots 51A to 51C is imaged from above by the camera 53, the position of each identification seal 52 in the image based on the obtained image signal S30 is detected, and the detection result And the color pattern information of each identification sticker 52 stored in the memory 57, the ID and position of each identification sticker 52 in the image are detected, and the detection result is transmitted to each robot 51A to 51C in the region 2. By doing so, the absolute positions of the robots 51A to 51C moving in the area 2 can be detected with high accuracy, and thus the robots 51A to 51C in the action area 2 can be reliably identified. An individual identification system and a robot having a configuration can be realized.

(3) Third Embodiment (3-1) Overall Configuration of Position Detection System According to Third Embodiment FIG. 13 where the same reference numerals are assigned to corresponding parts to FIG. 2 (A) is a position to which the present invention is applied. A detection system 70 is shown, and a wall 72 having a predetermined height is provided along the periphery of the action area 2 of the robot 71.

  In this case, the inner wall surface 72A of the wall 72 is painted with a different color for each of the wall surfaces 72AA to 72AD along each side of the region 2, and based on the painted color, the wall surface 72AA is the wall surface 72AA. It can be easily discriminated whether it is ˜72 AD.

  On the other hand, the robot 71 is shown in FIG. 15 in place of the individual identification unit 30 (FIG. 4) in the control unit 19 (FIG. 3) of the robots 3A to 3C (FIGS. 2A and 3) of the first embodiment. The robot 3A-3C is configured in the same manner as the robot 3A of the first embodiment except that a wall identifying unit 80 is provided.

  In this case, the wall identification unit 80 is provided with color extraction units 31A to 31D as many as the number of wall surfaces 72AA to 72AD along the sides of the region 2 (four surfaces), and the image signal S1 supplied from the camera 16 is provided. Is input to each of the color extraction units 31A to 31D.

  As a result, from each of the color extraction units 31A to 31D, a portion corresponding to a pixel of a corresponding color among pixels in the image based on the image signal S1 rises to a logic “1” level, and corresponds to a pixel of another color. The color extraction signals S10A to S10D whose part falls to the logic “0” level are output and supplied to the wall detection unit 81. Each of the color extraction units 31A to 31D extracts only one different color from among a plurality of colors for identification applied to the wall surfaces 72AA to 72AD.

  The wall detection unit 81 superimposes and scans images based on the color extraction signals S10A to S10D respectively supplied from the color extraction units 31A to 31D, so that a substantially horizontal, elongated region of the same color is set to any one of the wall surfaces 72AA to 72AD. And the color applied to the wall surfaces 72AA to 72AD is detected, and the height of the wall surfaces 72AA to 72AD in the image based on the image signal S1 is detected in units of pixels, and the obtained wall surfaces are detected. The colors 72AA to 72AD and the height in the image are sent to the comparison / calculation unit 82 as a wall detection signal S40.

  The comparison / calculation unit 82 uses the color of the wall surfaces 72AA to 72AD obtained on the basis of the wall detection signal S40, and a table relating to the colors and IDs of the wall surfaces 72AA to 72AD stored in the first memory 83 in advance. Based on this, the IDs of the wall surfaces 72AA to 72AD are searched.

  Further, the comparison / calculation unit 82 obtains the height (Hpic) in the image of the wall surfaces 72AA to 72AD obtained based on the wall detection signal S40 and a reference value (robot) stored in advance in the second memory 84. Based on the height (referred to as Hstd) in units of pixels of the wall surfaces 72AA to 72AD in the image when 71 is 1 [m] away from the wall surfaces 72AA to 72AD,

To calculate the distance L2 from the camera 16 to the wall surfaces 72AA to 72AD.

  Further, the comparison / calculation unit 82 then directs the camera 16 toward a wall surface (hereinafter referred to as a second wall surface) 72AA to 72AD different from its wall surface (hereinafter referred to as a first wall surface) 72AA to 72AD. Then, the same processing is executed to calculate the IDs of the second wall surfaces 72AA to 72AD and the distance L3 to the second wall surfaces 72AA to 72AD.

  Further thereafter, the comparison / calculation unit 82 calculates the ID of the first wall surfaces 72AA to 72AD obtained as described above, the distance L2 to the first wall surfaces 72AA to 72AD, and the second wall surfaces 72AA to 72AD. Map information of the area 2 including the ID and the distance L3 to the second wall surfaces 72AA to 72AD and the positions of the wall surfaces 72AA to 72AD along each side of the area 2 stored in advance in the third memory 85 Based on the above, it detects its own position in the area 2 and sends it as a position detection signal S41 to the uppermost CPU 36 that controls the behavior of the robot 71.

  As a result, in this robot 71, the CPU 36 can recognize its own position in the region 2 based on the position detection signal S41, and can act autonomously according to the surrounding situation based on the recognition result. It is made like that.

(3-2) Operation and Effect of Third Embodiment In the above configuration, in this position detection system 70, the robot 71 is based on the image signal S1 output from the camera 16, and the surrounding first and second wall surfaces. The color applied to 72AA to 72AD and the height Hpic of the first and second wall surfaces 72AA to 72AD in the image based on the image signal S1 are detected, and the detection result and the first memory 83 are stored in advance. An area based on the color and the ID of each wall surface 72AA to 72AD stored therein, the reference value Hstd stored in advance in the second memory 84, and the map information stored in advance in the third memory 85 2 to detect its own position.

  Therefore, in this position detection system 70, the robot 71 can easily and accurately recognize its own position in the region 2 from the colors of the two wall surfaces 72AA to 72AD and the height Hpic in the images of the wall surfaces 72AA to 72AD. it can.

  Moreover, in this position detection system 70, since it is only necessary to paint each wall surface 72AA-72AD arrange | positioned along each edge | side of the area | region 2 with a mutually different color, a system can be constructed | assembled very easily.

  Further, since this position detection system 70 does not use a method of transmitting a special signal such as a radio wave, it can be used without considering the influence on peripheral devices, radio wave regulations and the like. Since there is no need to install a signal generator on the surface, there is no possibility of hindering the movement of the robot 71.

  Furthermore, since this position detection system 70 is not a method of notating a symbol, a sign, or the like on the floor surface of the action area 2 of the robot 71, the floor surface can be painted for other purposes.

  Furthermore, in this position detection system 70, since the camera 16 of the robot 71 is oriented approximately horizontally, the wall surfaces 72AA to 72AD in front of the robot 71 can be imaged by the camera 16, so There is an advantage that the camera 16 does not need to be directed in a predetermined direction in order to detect the position of the camera, and that the user's own position in the region 2 can be detected while the other robot 71 is captured by the camera 16.

  According to the above configuration, the wall surfaces 72AA to 72AD are provided along the sides of the action area 2 of the robot 71 and different colors are applied to the wall surfaces 72AA to 72AD, respectively, while the robot 71 outputs from the camera 16. The color of at least two or more wall surfaces 72AA to 72AD imaged by the camera 16 and the height Hpic of each wall surface 72AA to 72AD in the image based on the image signal S1 are detected based on the image signal S1 to be performed. Then, the detection result, the identification color and the ID of each wall surface 72AA to 72AD stored in advance in the first memory 83, the reference value Hstd stored in advance in the second memory 84, the third By detecting the position of the user in the area 2 based on the map information of the area 2 stored in the memory 85 in advance, the robot DOO 71 can be detected accurately its position in the region 2, thus possible to realize a position detection system and a robot capable of accurately detecting the position of itself in the region 2.

(4) Fourth Embodiment (4-1) Principle In general, a color can be expressed by using three attributes of hue (Hue), saturation (Saturation) and lightness (Intensity).

  In this case, as shown in FIGS. 16A and 16B, the relationship between the hue, saturation (also referred to as saturation), and brightness is as follows. It can be expressed in polar coordinates with the angle around the origin O in this plane, the saturation as the distance from the origin O in this plane, and the brightness as the distance from the origin O in the direction perpendicular to this plane. Note that each vertex of the hexagon shown in FIG. 16B corresponds to R (red), Y (yellow), G (green), C (cyan), B (blue), and M (magenta), respectively.

  By the way, for example, as in the first to third embodiments, the color difference applied to the object (identifier 23 in the first embodiment, identification mark 52 in the second embodiment, and wall surfaces 72AA to 72AD in the third embodiment). In the system for identifying an object by the color, the appearance of the color changes depending on a change in illumination conditions, the direction in which the object is viewed, and the like even with the same color. For this reason, for example, in the first to third embodiments, this change is dealt with by giving a range to the condition for determining the color.

  However, even if such measures are taken, depending on how the colors used to paint the object are selected, when the lighting conditions change, the appearance of one color captured by the robot's vision falls within the criteria for judging another color. May be recognized as a different color, and as a result, an object painted with that color may not be correctly identified.

  As one of the methods for preventing such a situation, for example, as a combination of colors used for identification, colors that are separated from each other in a predetermined color space are used in advance, and are output from the camera inside the robot. The image signal format may be converted into the color space format to perform color identification.

  That is, as the identification colors, for example, a plurality of colors whose hues are separated by a predetermined angle (for example, 60 °) or more in a color space (hereinafter, referred to as an HSI space) that uses three attributes of hue, saturation, and brightness. In addition to selecting a color, the image format of the image signal output from the camera inside the robot is a format that represents the color as hue (H), saturation (S), and brightness (I) (hereinafter referred to as the HSI format). ) And identifying the color from the hue based on the image signal thus obtained makes it less likely to be affected by changes in the lighting conditions during color identification, which makes it possible to efficiently determine the wrong color identification. It can be avoided well.

(4-2) Configuration of Position Detection System According to Fourth Embodiment FIG. 17 in which parts corresponding to those in FIG. 14 are assigned the same reference numerals shows a position detection system 90 according to the fourth embodiment. In addition, different unique colors are applied to the wall surfaces 72AA to 72AD of the wall 72 along each side of the region 2.

  In this case, five colors (for example, R, Y, G, C, and B) whose hues are separated by 60 [°] or more in the HSI space are selected as colors to be applied to the region 2 and the wall surfaces 72AA to 72AD. .

  On the other hand, in the robot 91, position detection as shown in FIG. 18 is performed by assigning the same reference numerals to the corresponding parts in FIG. 15 instead of the position detection unit 80 (FIG. 15) in the robot 71 (FIG. 14) of the third embodiment. The configuration is the same as that of the robot 71 except that the unit 92 is provided.

  In this case, the position detection unit 92 is provided with an image format conversion unit 93 and the same number of color extraction units 94A to 94D as the number of wall surfaces 72AA to 72AD. The image format conversion unit 93 is supplied from the camera 16. The image signal S1 is converted to an HSI format image signal S50, which is sent to each of the color extraction units 94A to 94D.

  Each of the color extraction units 94A to 94D extracts a pixel having a hue within a predetermined angle from the hue specified in the polar coordinates shown in FIG. 16B based on the hue information of each pixel included in the image signal S50. The portion corresponding to the pixel of the color to be extracted rises to the logical logic “1” level based on the detection result, and the portion corresponding to the pixel of the other color falls to the logical “0” level. The color extraction signals S51A to S51D are generated and sent to the wall detection unit 81. Each of the color extraction units 94A to 94D extracts pixels of predetermined colors different from each other from a plurality of colors applied to the wall surfaces 72AA to 72AD.

  As a result, as described above in FIG. 15, based on these color extraction signals S51A to S51D, the color of the wall surface 72AA to 72AD captured by the camera 16 and the pixels in the image of the wall surface 72AA to 72AD are used as a unit. The height hpic is detected by the wall detector 81, and based on the detection result, the comparison / calculation unit 82 detects its own position in the region 2 in the same manner as in the third embodiment. Is sent as a position detection signal S41 to the uppermost CPU 36 that controls the behavior of the robot 91.

  As a result, in this robot 91, the CPU 36 can recognize its own position in the area 2 based on the position detection signal S41, and can autonomously take action according to the surrounding situation based on the recognition result. It is made like that.

(4-3) Operation and Effect of Fourth Embodiment In the above configuration, the position detection system 90 outputs the robot 91 from the camera 16 as in the position detection system 70 (FIG. 14) of the third embodiment. Based on the image signal S1, the color of the surrounding first and second wall surfaces 72AA to 72AD and the height Hpic of the first and second wall surfaces 72AA to 72AD in the image based on the image signal S1 are detected. These detection results, the colors and IDs of the wall surfaces 72AA to 72AD stored in advance in the first memory 83, the reference value Hstd stored in advance in the second memory 84, and the third memory 85 Based on the map information stored in advance, the user's position in the area 2 is detected.

  Therefore, according to this position detection system 90, the same effect as the position detection system 70 of the third embodiment can be obtained.

  In addition to this, in this position detection system 90, the colors applied to each of the wall surfaces 72AA to 72AD of the wall 72 and the region 2 are colors whose hues are separated by 60 ° or more in the HSI space. Since the color identification is performed after the image format of the image signal S1 output from the camera 16 is converted to the HSI format as an internal process, the robot 91 can be used even when the illumination conditions for the wall surfaces 72AA to 72AD change. It is possible to make it less susceptible to the influence of color identification inside, and thus it is possible to efficiently avoid erroneous color identification determination in the robot 91.

  According to the above configuration, with respect to the position detection system 70 (FIG. 14) of the third embodiment, the hue in the HSI space is 60 [°] as the color applied to each of the wall surfaces 72AA to 72AD of the wall 72 and the region 2. In addition to using colors that are separated from each other, the image signal S1 output from the camera 61 as an internal process of the robot 91 is converted into an image signal S50 in the HSI format, and then color identification is performed based on the image signal S50. Accordingly, it is possible to efficiently avoid erroneous determination of color identification in the robot 91. Accordingly, it is possible to realize a position detection system and a robot apparatus that can avoid erroneous recognition of an object inside the robot 91 and thus can detect the position of the user in the area with higher accuracy.

(5) Fifth Embodiment (5-1) Configuration of Position Detection System According to Fifth Embodiment FIG. 19 shows the position detection system 100 according to the fifth embodiment shown in FIG. For example, the wall surfaces 72AA to 72AD of the wall 72 are painted with different unique colors whose hues are separated by 60 ° or more in the HSI space.

In this case, as shown in FIG. 20, the color of each of the wall surfaces 72AA to 72AD is lowest at one end in the longitudinal direction of each of the wall surfaces 72AA to 72AD, and the saturation (saturation) increases linearly from one end to the other end. That is, when the saturation of the color at one end of the wall surfaces 72AA to 72AD is S _min , the saturation at the other end is S _max and the length in the longitudinal direction of the wall surfaces 72AA to 72AD is M, the wall surface 72AA The saturation S _x of the color at a distance x from one end of 72 AD is

Painted to satisfy.

  On the other hand, in the robot 101, a position detecting unit as shown in FIG. 21 in which the same reference numerals are assigned to corresponding parts in FIG. 18 instead of the position detecting unit 92 (FIG. 18) in the robot 91 (FIG. 17) of the fourth embodiment. The robot is configured in the same manner as the robot 91 except that 102 is provided.

  In this case, the position detection unit 102 is provided with the same number of color extraction units 94A to 94D as the number of wall surfaces 72AA to 72AD along each side of the region 2, and the image format of the image signal S1 output from the camera 16 is changed. After the image format conversion unit 93 converts the image signal into the HSI format, the image signal S50 thus obtained is sent to each of the color extraction units S51A to S51D.

  In each of the color extraction units S51A to S51D, a pixel of a predetermined color is extracted from the image based on the image signal S50, and a portion corresponding to the pixel rises to a logical “1” level and corresponds to a pixel of any other color. The color extraction signals S51A to S51D that fall to the logic “0” level are sent to the wall detection unit 103, respectively.

  The wall detection unit 103 superimposes and scans images based on the color extraction signals S51A to S51D supplied from the color extraction units 94A to 94D, respectively, so that a substantially horizontal elongated region of the same color is placed on any one of the wall surfaces 72AA to 72AD. And the detection result is given as a wall detection signal S60 to the uppermost CPU 36 that controls the behavior of the robot.

  At this time, the CPU 36 moves the direction of the camera 16 (that is, the head of the robot 101) in the left-right direction by driving the actuator of the corresponding joint portion, and the wall detection signal S60 supplied from the wall detection unit 103 is detected. Based on this, the wall surface 72AA to 72AD having the largest number of pixels in the image based on the image signal S50 (that is, the closest wall surface) 72AA to 72AD is detected, and the upper or lower end in the image of the wall surface 72AA to 72AD is horizontal ( That is, the orientation of the camera 16 is adjusted so that the optical axis of the camera 16 is perpendicular to the wall surfaces 72AA to 72AD.

In the state where the optical axis of the camera 16 is aligned perpendicularly to the closest wall surfaces 72AA to 72AD, the wall detection unit 103 is based on the color extraction signals S51A to S51D respectively supplied from the color extraction units 94A to 94D. The color of the wall surfaces 72AA to 72AD is detected and applied to the wall surfaces 72AA to 72AD at the center of the image based on the image signal S50 based on the image signal S50 supplied from the image format conversion unit 93 at this time. The color saturation _Sx is detected, and the detected color of the wall surfaces 72AA to 72AD and the saturation _{Sx of} the color applied to the wall surfaces 72AA to 72AD in the center of the image are detected as the color and saturation detection signal S61. To the comparison / calculation unit 104.

  The comparison / calculation unit 104 is a table regarding the colors of the wall surfaces 72AA to 72AD obtained based on the color and saturation detection signal S61, the colors of the wall surfaces 72AA to 72AD stored in the first memory 105, and their IDs. And the IDs of the wall surfaces 72AA to 72AD are detected.

The comparison and calculation unit 104, and saturation _{S x} of the wall 72AA～72AD in the image central portion obtained based on the color and the saturation signal S61, the wall surfaces are previously stored in the second memory 106 72AA～72AD By calculating backwardly the equation (3) based on the saturation S _min and S _max at one end and the other end of the S and the length M of each of the wall surfaces 72AA to 72AD, the user's position in the direction parallel to the wall surfaces 72AA to 72AD Is detected.

  Further, the comparison / calculation unit 105 then rotates the camera 16 by 90 [°] under the control of the CPU 36, and the wall surface is different from the wall surfaces 72AA to 72AD (hereinafter referred to as the first wall surfaces 72AA to 72AD). 72AA to 72AD (hereinafter referred to as second wall surfaces 72AA to 72AD), and then the same processing is performed to execute the second wall surfaces 72AA to 72AD and the second wall surfaces 72AA to 72AA. It detects its position in a direction parallel to 72AD.

  Further, the comparison / calculation unit 104 then obtains the IDs of the first wall surfaces 72AA to 72AD obtained as described above, the positions in the direction parallel to the first wall surfaces 72AA to 72AD, and the second wall surfaces 72AA to 72AA. Based on the ID of 72AD and the position in the direction parallel to the second wall surfaces 72AA to 72AD and the map information of the region 2 including the positions of the wall surfaces 72AA to 72AD stored in advance in the third memory 107 The position of itself is detected, and the detection result is sent to the CPU 36 as a position detection signal S62.

  As a result, in this robot 101, the CPU 36 can recognize its own position in the region 2 based on the position detection signal S62, and can autonomously take action according to the surrounding situation based on the recognition result. It is made like that.

(5-2) Operation and Effect of Fifth Embodiment In the above configuration, in the position detection system 100, the robot 101 is based on the image signal S1 output from the camera 16 and the surrounding first and second wall surfaces 72AA. ˜72AD and the saturation _{Sx at} the center of the image based on the image signal S1 of these colors are detected, and these detection results and the colors of the wall surfaces 72AA to 72AD stored in the first memory 105 are detected. And the ID thereof, the length M of each of the wall surfaces 72AA to 72AD stored in the second memory 106, and the color saturation S _min and S _max at one end and the other end of each of the wall surfaces 72AA to 72AD, Based on the map information stored in the memory 107, the user's position in the area 2 is detected.

  Therefore, in this position detection system 100, the robot 101 can easily and accurately recognize its own position in the region 2 based on the change in the color of the two wall surfaces 72AA to 72AD and the saturation of these colors.

  Further, in this position detection system 100, it is only necessary to apply different colors to the wall surfaces 72AA to 72AD arranged along each side of the region 2 while changing the saturation in the horizontal direction. Can be built.

  Furthermore, since this position detection system 100 does not use a method for transmitting a special signal such as a radio wave, the position detection system 100 can be used without considering the influence on peripheral devices, radio wave regulations, and the like. Since there is no need to install a signal generator, there is no possibility of disturbing the behavior of the robot 101.

  Furthermore, since this position detection system 100 is not a method of writing symbols, signs, etc. on the floor surface of the action area 2 of the robot 101, the floor surface can be painted for other purposes.

  Furthermore, in this position detection system 100, since the camera 16 of the robot 101 is oriented approximately horizontally, the wall surfaces 72AA to 72AD in front of the robot 101 can be imaged by the camera 16, so There is no need to point the camera 16 in a predetermined direction in order to detect the position of the camera, and there is an advantage that it is possible to detect its own position in the region 2 while capturing other robots with the camera 16.

According to the above configuration, different colors are applied to the respective wall surfaces 72AA to 72AD while changing the saturation in the horizontal direction, and the robot 102 applies the colors of the two wall surfaces 72AA to 72AD and these wall surfaces 72AA to 72AD. When the image is captured vertically, the saturation _Sx at the center of the image of the color is detected, and the detection result, the color of each of the wall surfaces 72AA to 72AD stored in the first memory 105, and its ID , The length M of each of the wall surfaces 72AA to 72AD stored in the second memory 106, and the color saturations S _min and S _max at one end and the other end of each of the wall surfaces 72AA to 72AD, and the third memory 107 Since the robot 101 detects its own position in the area 2 based on the map information that has been recorded, the robot 101 can detect its position in the area 2. It is possible to realize a position detection system and a robot that can detect the position with high accuracy and thus can detect the position of the user in the region 2 with high accuracy.

(6) Sixth Embodiment (6-1) Configuration of Position Detection System According to Sixth Embodiment FIG. 22 showing the same reference numerals in FIG. 19 corresponding to those in FIG. 19 shows a position detection system 110 according to the sixth embodiment. For example, different unique colors whose hues are separated by 60 ° or more in the HSI space are applied to the wall surfaces 72AA to 72AD of the wall 72, respectively.

  Each of the wall surfaces 72AA to 72AD has a color separated from the hue of each color applied to each of the wall surfaces 72AA to 72AD in the HSI space by 60 [°] or more from the vicinity of the lower end of one end to the vicinity of the upper end of the other end. A hatched line 102 is used.

  On the other hand, in the robot 111, the position detection as shown in FIG. 24 is performed by assigning the same reference numerals to the robot 91 (FIG. 17) of the fourth embodiment in place of the position detection unit 92 (FIG. 18). The configuration is the same as that of the robot 91 except that the unit 113 is provided.

  In this case, the position detection unit 113 is provided with one more color extraction units 94A to 94E than the number of wall surfaces 72AA to 72AD along each side of the region 2, and the image signal S1 output from the camera 16 is provided. After the image format is converted into the HSI format by the image format conversion unit 93, the image signal S50 thus obtained is sent to each of the color extraction units 94A to 94E.

  In each of the color extraction units 94A to 94E, a pixel of a predetermined color corresponding to each of the images based on the image signal S50 is extracted, and a portion corresponding to the pixel rises to a logic “1” level. The color extraction signals S51A to S51E in which the portion corresponding to falls to the logic “0” level are sent to the wall detection unit 114, respectively. Each color extraction unit 114 extracts pixels having different colors from the color applied to each of the wall surfaces 72AA to 72AD and the shaded line 112.

  The wall detection unit 114 superimposes and scans images based on the color extraction signals S51A to S51E supplied from the color extraction units 94A to 94E, respectively, so that one of the wall surfaces 72AA to 72AD is aligned with a substantially horizontal, elongated region of the same color. And the detection result is given as a wall detection signal S60 to the uppermost CPU 36 that controls the behavior of the robot 111.

  At this time, the CPU 36 moves the direction of the camera 16 (that is, the head of the robot 111) in the left-right direction by driving the actuators of the corresponding joints, and the wall detection signal S60 supplied from the wall detection unit 114 is detected. Based on this, the wall surface 72AA to 72AD having the largest number of pixels in the image based on the image signal S50 (that is, the closest wall surface) 72AA to 72AD is detected, and the upper or lower end in the image of the wall surface 72AA to 72AD is horizontal ( That is, the orientation of the camera 16 is adjusted so that the optical axis of the camera 16 is perpendicular to the nearest wall surfaces 72AA to 72AD.

In the state in which the optical axis of the camera 16 is aligned perpendicularly to the nearest wall surfaces 72AA to 72AD, the wall detection unit 114 is longer in the vertical direction in the upper part than the oblique line 112 of the wall surfaces 72AA to 72AD in the center of the image. The length U _x and the length L _x below the oblique line 112 are detected in units of pixels, and the detected length U _x and the lower part of the wall portion 72AA to 72AD above the oblique line 112 are detected. a length _{L x,} and sends the comparison and calculation section 115 and the color of the walls 72AA~72AD as the color and length detection signal S70.

  The comparison / calculation unit 115 is a table relating to the color of the wall surfaces 72AA to 72AD obtained based on the color and length detection signal S70, the color of each wall surface 72AA to 72AD stored in the first memory 116 and the ID thereof. And the IDs of the wall surfaces 72AA to 72AD are detected.

The comparison and calculation unit 115, the color and length detection signal obtained based on the S70 than hatching 112 in the wall 72AA~72AD based on the length _{L x} of length _{U x} and the lower portion of the upper part following formula

The vertical length U _x of the portion above the oblique line 112 with respect to the height (= U _x + L _x ) of the wall surfaces 72AA to 72AD at the center of the image based on the image signal S50 to calculate the ratio of _{R x1.}

The comparison / calculation unit 115 then calculates the calculation result, the length M of each of the wall surfaces 72AA to 72AD stored in advance in the second memory 117, and the height of the wall surfaces 72AA to 72AD at one end of each of the wall surfaces 72AA to 72AD. The ratio R _a1 of the length in the vertical direction of the portion above the oblique line 112 to the height R and the ratio R _b1 of the length of the portion above the oblique line 112 to the height of the wall surface 72AA to 72AD at the other end of each wall surface 72AA to 72AD Based on

The distance x from the one end of the wall surfaces 72AA to 72AD to the portion imaged at the center of the image of the wall surfaces 72AA to 72AD (the wall surface in a direction parallel to the wall surfaces 72AA to 72AD) 72AA to 72AD) (corresponding to the distance from the robot 111 to one end).

It should be noted that the ratio R _a1 of the length of the portion above the oblique line 112 to the height of the wall surfaces 72AA to 72AD at one end of each wall surface 72AA to 72AD is that at one end of each of the wall surfaces 72AA to 72AD, as shown in FIG. than hatching 112 wall 72AA~72AD the length of the upper part and _{U a1,} the following equation the length of the lower part as _{L a1} than hatching 112

Given by the ratio _{R b1} of the length of the upper part than the hatched 112 to the height of the wall 72AA~72AD at the other end of each wall 72AA~72AD, the wall 72AA~ at the other end of the wall surface 72AA~72AD The length of the upper part of the 72AD diagonal line 112 is U _b1, and the length of the lower part of the diagonal line is L _b1.

Given by.

  Further, the comparison / calculation unit 115 then rotates the camera 16 by 90 [°] under the control of the CPU 36, so that the wall surface is different from the wall surfaces 72AA to 72AD (hereinafter referred to as the first wall surfaces 72AA to 72AD). 72AA to 72AD (hereinafter referred to as second wall surfaces 72AA to 72AD), and then the same processing is performed to execute the second wall surfaces 72AA to 72AD and the second wall surfaces 72AA to 72AA. It detects its position in a direction parallel to 72AD.

  Further, the comparison / calculation unit 115 then obtains the IDs of the first wall surfaces 72AA to 72AD obtained as described above, the positions in the direction parallel to the first wall surfaces 72AA to 72AD, and the second wall surfaces 72AA to 72AA. Based on the ID of 72AD and the position in the direction parallel to the second wall surfaces 72AA to 72AD, and the map information of the region including the position of each wall surface 72AA to 72AD stored in advance in the third memory 118 The position of itself is detected, and the detection result is sent to the CPU 36 as a position detection signal S71.

  As a result, in this robot 111, the CPU 36 can recognize its own position in the region 2 based on the position detection signal S71, and can automatically take action according to the surrounding situation based on the recognition result. It is made like that.

(6-2) Operation and Effect of Sixth Example In the above configuration, in the position detection system 110, the robot 111 is based on the image signal S1 output from the camera 16 and the surrounding first and second wall surfaces 72AA. ˜72AD and the ratio R _x1 of the length U _x of the portion above the oblique line 112 with respect to the height of each of the wall surfaces 72AA to 72AD in the center of the image based on the image signal S1, and the detection results The color and the ID of each wall surface 72AA to 72AD stored in advance in the first memory 116, the length M of each wall surface 72AA to 72AD stored in advance in the second memory 117, and the wall surfaces 72AA to 72AA. the length of the upper part than the hatched 112 to the height of the wall 72AA~72AD at one end and the other end of 72AD _{U a1,} the ratio of _{U b1 R a 1} and R _b1 and the map information stored in advance in the third memory 118 to detect its own position in the area 2.

  Therefore, in this position detection system 110, the robot 111 can easily and accurately recognize its own position in the area 2 based on the colors of the two wall surfaces 72AA to 72AD and the oblique lines 112 written on these wall surfaces 72AA to 72AD. it can.

  In addition, in this position detection system 110, it is only necessary to apply different colors to the wall surfaces 72AA to 72AD arranged along each side of the region 2 and to indicate the oblique lines 112, so that the system can be constructed very easily. can do.

  Further, since the position detection system 110 does not use a method of transmitting a special signal such as a radio wave, the position detection system 110 can be used without considering the influence on peripheral devices, radio wave regulations, etc. Since it is not necessary to install a signal generator on the surface, there is no possibility of hindering the behavior of the robot 111.

  Furthermore, since this position detection system 110 is not a method of writing symbols, signs, etc. on the floor surface of the action area 2 of the robot 111, the floor surface can be painted for other purposes.

  Further, in this position detection system 110, since the camera 16 of the robot 111 is oriented approximately in the horizontal direction, the front wall surfaces 72AA to 72AD of the robot 111 can be imaged by the camera 16, so There is no need to point the camera 16 in a predetermined direction in order to detect the position of the camera, and there is an advantage that it is possible to detect its own position in the region 2 while capturing other robots with the camera 16.

According to the above configuration, different colors are applied to the respective wall surfaces 72AA to 72AD, and the diagonal lines 112 are written on the respective wall surfaces 72AA to 72AD, while the robot 111 has the colors of the two wall surfaces 72AA to 72AD and the wall surfaces. When the vertical images of 72AA to 72AD are captured, the ratio R _x1 of the longitudinal length U _x above the oblique line 112 at the center of the image is detected, and these detection results and the first memory 116 are detected. The color and the ID of each wall surface 72AA to 72AD stored in advance, the length M of each wall surface 72AA to 72AD stored in advance in the second memory 117, and the one at the one end and the other end of each wall surface 72AA to 72AD the ratio _{R a1, R b1} of the upper part of the length _{U a1, U b1} than hatching 112 to the height of the wall 72AA～72AD, the By detecting the position of the robot in the area 2 based on the map information stored in advance in the memory 118, the robot 111 can accurately detect the position of the robot in the area 2, Thus, a position detection system and a robot that can accurately detect the position of the user in the area 2 can be realized.

(7) Seventh Embodiment (7-1) Configuration of Position Detection System According to Seventh Embodiment FIG. 25, in which parts corresponding to those in FIG. 14 are assigned the same reference numerals, shows a position detection system 120 according to the seventh embodiment. For example, colors of different hues separated by 60 [°] or more in the HSI space are assigned to the wall surfaces 72AA to 72AD of the wall 72, respectively.

  That is, as shown in FIG. 26, each of the wall surfaces 72AA to 72AD has a virtual line K1 extending from the vicinity of the lower end of one end in the longitudinal direction to the vicinity of the upper end portion of the other end as a boundary from the virtual line K1 of each wall surface 72AA to 72AD. Also, a color with a high hue and a predetermined hue is applied to the upper part, and a color with a low saturation and the same hue as the upper part is applied to the lower part of the virtual line K1.

  On the other hand, in the robot 121, the position detection as shown in FIG. 27, in which the same reference numerals are given to the corresponding parts in FIG. 18 instead of the position detection unit 92 (FIG. 18) in the robot 91 (FIG. 17) of the fourth embodiment. The configuration is the same as that of the robot 91 except that the unit 122 is provided.

  In this case, the position detection unit 122 is provided with the same number of color extraction units 94A to 94D as the number of wall surfaces 72AA to 72AD along each side of the region 2, and the image format of the image signal S1 output from the camera 16 is provided. Is converted into the HSI format by the image format conversion unit 93, and the image signal S50 thus obtained is sent to the color extraction units 94A to 94D and the wall detection unit 123, respectively.

  As described above, each of the color extraction units 94A to 94D selects pixels having a hue within a predetermined angle from the hue specified in the polar coordinates shown in FIG. 16B based on the hue information of each pixel included in the image signal S50. A portion corresponding to a pixel of a color to be detected is detected as a pixel of a color to be extracted and based on the detection result, and a portion corresponding to a pixel of another color rises to a logic “0” level. The color extraction signals S51A to S51D that fall on are generated and sent to the wall detection unit 123.

  The wall detection unit 123 also superimposes and scans images based on the color extraction signals S51A to S51D respectively supplied from the color extraction units 94A to 94D, so that a substantially horizontal elongated region having the same hue is set to any one of the wall surfaces 72AA. ˜72AD, and the detection result is given as a wall detection signal S80 to the uppermost CPU 36 that controls the action of the robot 121.

  At this time, the CPU 36 moves the direction of the camera (that is, the head of the robot) in the left-right direction by driving the actuator of the corresponding joint, and based on the wall detection signal S80 supplied from the wall detection unit 123. The wall surfaces 72AA to 72AD having the largest number of pixels in the image based on the image signal S50 (that is, the closest wall surfaces) 72AA to 72AD are detected, and the upper or lower ends in the images of the wall surfaces 72AA to 72AD are horizontal (that is, The orientation of the camera 16 is adjusted so that the optical axis of the camera 16 is perpendicular to the wall surfaces 72AA to 72AD.

Then, in a state where the optical axis of the camera 16 is oriented perpendicularly to the closest wall surfaces 72AA to 72AD, the wall detection unit 123 receives the color extraction signals S51A to S51D given from the color extraction units 94A to 94D and the image format conversion unit 93. Based on the saturation information of each pixel included in the given image signal S50, a portion to which the color having the higher saturation of the wall surfaces 72AA to 72AD in the central portion of the image based on the image signal S50 is applied (ie, virtual The portion above the line K1, hereinafter referred to as the high saturation portion of the wall surfaces 72AA to 72AD) in the vertical direction H _x (FIG. 26) and the smaller one of the saturation of the wall surfaces 72AA to 72AD at the center of the image Of the portion to which the color is applied (that is, the portion below the imaginary line K1, hereinafter referred to as the low saturation portion of the wall surfaces 72AA to 72AD) Length _{L x} (Fig. 26) and the respectively detected pixels as a unit, the length _{H x} of the high saturation of these detected walls 72AA～72AD length _{L x} of the low chroma portion of the wall surface 72AA～72AD The hues of the wall surfaces 72AA to 72AD are sent to the comparison / calculation unit 124 as a wall surface detection signal S81.

  The comparison / calculation unit 124 is based on the hues of the wall surfaces 72AA to 72AD obtained based on the wall surface detection signal S81, and the hue and the IDs of the wall surfaces 72AA to 72AD stored in the first memory 125 in advance. Then, the IDs of the wall surfaces 72AA to 72AD are detected.

The comparison and calculation unit 124, the lengths _H x of the high saturation portions and low chroma portion of the wall surface 72AA~72AD in the image central portion obtained on the basis of a wall surface detection signal S81, on the basis of the _{L x} following formula

The ratio R _x2 of the vertical length of the high saturation portion with respect to the height of the wall surface at the center portion of the image calculated in this way, and each wall surface stored in advance in the second memory 126 The ratio R _a2 of the length of the high saturation portion to the height of the wall surfaces 72AA to 72AD at one end of 72AA to 72AD, and the ratio of the length of the high saturation portion to the height of the wall surfaces 72AA to 72AD at the other end of each wall surface 72AA to 72AD based on the length of R _b2 and each wall 72AA~72AD M, the following equation

The distance x from the one end of the wall surfaces 72AA to 72AD to the portion imaged at the center of the image of the wall surfaces 72AA to 72AD (the wall surface in a direction parallel to the wall surfaces 72AA to 72AD) 72AA to 72AD) (corresponding to the distance from the robot 121 to one end).

The ratio R _a2 of the longitudinal length of the high saturation portion to the height of the wall surfaces 72AA to 72AD at one end of each of the wall surfaces 72AA to 72AD is as shown in FIG. _Assuming that the length is H _a2 and the length of the low saturation portion is _La2 ,

Given by. Further, the ratio R _b2 of the length of the high saturation portion to the height of the wall surface 72AA to 72AD at the other end of each wall surface 72AA to 72AD is set to H _b2 with the length of the high saturation portion at the other end of the wall surface 72AA to 72AD being low. The length of the saturation part is L _b2 and the following formula

Given by.

  Thereafter, the comparison / calculation unit 124 is different from the wall surfaces 72AA to 72AD (hereinafter referred to as the first wall surfaces 72AA to 72AD) when the camera 16 is rotated by 90 [°] under the control of the CPU 36. 72AA to 72AD (hereinafter referred to as second wall surfaces 72AA to 72AD), and then the same processing is performed to execute the second wall surfaces 72AA to 72AD and the second wall surfaces 72AA to 72AA. It detects its position in a direction parallel to 72AD.

  Further, the comparison / calculation unit 124 then obtains the IDs of the first wall surfaces 72AA to 72AD obtained as described above, the positions in the direction parallel to the first wall surfaces 72AA to 72AD, and the second wall surfaces 72AA to 72AA. Based on the ID of 72AD and the position in the direction parallel to the second wall surfaces 72AA to 72AD and the map information of the region including the positions of the wall surfaces 72AA to 72AD stored in advance in the third memory 127 The position of itself is detected, and the detection result is sent to the CPU 36 as a position detection signal S82.

  As a result, in this robot 121, the CPU 36 can recognize its own position in the region 2 based on the position detection signal S82, and can automatically take action according to the surrounding situation based on the recognition result. It is made like that.

(7-2) Operation and Effect of Seventh Example In the above configuration, in the position detection system 120, the robot 121 is based on the image signal S1 output from the camera 16, and the surrounding first and second wall surfaces 72AA. ˜72AD and a ratio R _x2 of the length H _x of the high saturation portion to the height of each of the wall surfaces 72AA to 72D in the center portion of the image based on the image signal S1, and the detection result and the first The color and the ID of each wall surface 72AA to 72AD stored in the memory 125, the length M of each wall surface 72AA to 72AD stored in the second memory 126, and one end and the other end of each wall surface 72AA to 72AD high saturation portion with respect to the height of the wall 72AA~72AD the ratio _{R a2, R b2} length _{H a2, H b2,} are stored in the third memory 127 To detect its own position within the region 2 on the basis of the FIG information.

  Therefore, in this position detection system 120, the robot 121 can easily and accurately recognize its own position in the region 2 based on the hues of the two wall surfaces 72AA to 72AD and the high saturation portions of these wall surfaces 72AA to 72AD. .

  Further, in this position detection system 120, it is only necessary to apply two colors having different hues and greatly different saturations to each of the wall surfaces 72AA to 72AD arranged along each side of the region 2 in a predetermined pattern. A system can be constructed very easily.

  Further, since the position detection system 120 does not use a method of transmitting a special signal such as a radio wave, the position detection system 120 can be used without considering the influence on peripheral devices, radio wave regulations, and the like. Since it is not necessary to install a signal generator on the surface, there is no possibility of hindering the behavior of the robot 121.

  Furthermore, since this position detection system 120 is not a method of writing symbols, signs, or the like on the floor surface of the action area 2 of the robot 121, the floor surface can be painted for other purposes.

  Furthermore, in this position detection system 120, the camera 121 of the robot 121 can be imaged by the camera 16 by directing the camera 16 of the robot 121 in a substantially horizontal direction. There is no need to point the camera 16 in a predetermined direction in order to detect the position of the camera, and there is an advantage that it is possible to detect its own position in the region 2 while capturing other robots with the camera 16.

According to the above configuration, two different colors with the same hue and saturation are applied to each of the wall surfaces 72AA to 72AD in a predetermined pattern using different hues for each of the wall surfaces 72AA to 72AD. detection and hue of the wall 72AA～72AD, vertical high saturation portion with respect to the height of the wall 72AA～72AD these walls 72AA～72AD the image central portion in the case where the captured vertically by the camera 16 the length and _{H x,} respectively These detection results, the colors and IDs of the wall surfaces 72AA to 72AD stored in the first memory 125, the lengths M of the wall surfaces 72AA to 72AD stored in the second memory 126, and the respective IDs. length of the high saturation portions with respect to the height of the wall 72AA～72AD at one end and the other end of the wall surface 72AA~72AD _{H a2, b2} the ratio R _a2, R _b2, and by which is adapted to detect its position in the third on the basis of the map information stored in the memory 127 in the region 2, own robot 121 is in the area 2 Thus, it is possible to realize a position detection system and a robot that can accurately detect the position of the robot 2 and thus can accurately detect the position of the user in the region 2.

(8) Eighth Embodiment (8-1) Configuration of Position Detection System According to Eighth Embodiment FIG. 28, in which parts corresponding to those in FIG. 17 are assigned the same reference numerals, shows a position detection system 130 according to the eighth embodiment. The position detection system 90 of the fourth embodiment is shown except that the wall surfaces 131AA to 131AD provided along the sides of the region 2 are respectively constituted by the panel surfaces of the plurality of liquid crystal panels 132. The configuration is the same as in FIG.

  In this case, as shown in FIG. 29, the liquid crystal panels 132 having the same color along the same side of the region 2 and the hues along the different sides of the region 2 have a hue of 60 ° or more in the HSI space. Control is performed by the control unit 133 so that light of different colors separated from each other is emitted into the region 2.

  As a result, in this position detection system 130, as in the position detection system 90 (FIG. 17) of the fourth embodiment, the robot 91 easily identifies the wall surfaces 131AA to 131AD based on the colors of the wall surfaces 131AA to 131AD. Has been made to get.

(8-2) Operation and Effect of Eighth Embodiment In the above configuration, in this position detection system 130, as described above with reference to FIG. 18, the robot 91 is operated based on the image signal S1 output from the camera 16. The color of the first and second wall surfaces 131AA to 131AD and the height of the first and second wall surfaces 131AA to 131AD in the image based on the image signal S1 are detected, the detection result and the first memory 83, the color of each of the wall surfaces 131AA to 131AD and the ID thereof, the reference value Hstd stored in advance in the second memory 84, and the map information stored in advance in the third memory 85. Based on this, the user's position in the area 2 is detected.

  Therefore, according to this position detection system 130, the same effect as the position detection system 90 (FIG. 17) of the fourth embodiment can be obtained.

  In addition to this, in this position detection system 130, each of the wall surfaces 131AA to 131AD emits a predetermined color by itself, so that, for example, a robot is used as compared with the case where the reflection of light on each of the wall surfaces 72AA to 72AD is used as in the fourth embodiment. When performing color identification inside 91, it can be made less susceptible to the influence of the external environment such as lighting.

  Further, in the position detection system 130, the wall surfaces 131AA to 131AD are configured by the panel surface of the liquid crystal panel 132, so that the colors of the wall surfaces 131AA to 131AD can be freely set according to the type of the robot 91 and the work content. It also has the advantage that it can be switched.

  According to the above configuration, the wall surfaces 131AA to 131AD along the respective sides of the region 2 are configured by the panel surfaces of the plurality of liquid crystal panels 132, respectively, with respect to the position detection system 90 (FIG. 17) of the fourth embodiment. Those liquid crystal panels 132 that have the same color along the same side of the region 2 and those along the different side of the region 2 emit light of different colors whose hues are separated by 60 ° or more in the HSI space. By controlling so that it fires into the area 2, it is possible to make the color identification inside the robot 91 less susceptible to the influence of changes in the external environment, thus detecting its own position within the area 2 with high accuracy. A position detection system and a robot that can do this can be realized.

(9) Other Embodiments In the above-described first to eighth embodiments, the present invention is applied to the autonomous mobile robots 3A to 3C, 51A to 51C, 71, 91, 101, 111, 121. However, the present invention is not limited to this, and can be applied to various other robots or other moving bodies.

  In the first embodiment described above, the case where the identification body 23 is attached to each of the robots 3A to 3C via the support rod 22 as shown in FIG. 2A has been described. For example, the identification body 23 may be disposed on the head 12 of the robot 140 as shown in FIG. 30. The point is that the robot 3A to 3C is easily visible from the other robots 3A to 3C. If the identification body 23 is attached, various other positions can be applied as the attachment position of the identification body 23.

  Further, in the first embodiment described above, the case where the identification body 23 is formed in a spherical shape has been described. However, the present invention is not limited to this, for example, the shape of an elliptic rotating body as shown in FIG. Various other shapes such as a cylindrical shape as shown in FIG. However, the shape of the identification body 23 is selected as a rotating body having a central axis perpendicular to the moving direction of the robots 3A to 3C, and the surface is divided into a plurality of belt-like regions parallel to the moving direction of the robots 3A to 3C. By coloring each band-like region in a predetermined color with the color pattern of FIG. 5, the identification object can be seen with the same shape and the same color pattern when viewed from any direction when the region 2 is flat. Individual identification of the body can be facilitated.

  Furthermore, in the first embodiment described above, the case where the surface of the identification body 23 is color-coded is described. However, the present invention is not limited to this, and the color may be color-coded to other numbers. . In the first embodiment described above, the case where 16 colors are prepared for identification has been described. However, the present invention is not limited to this, and other numbers may be used.

  Furthermore, in the first embodiment described above, the camera 16 is applied as an imaging means for imaging the identification body 23 having a different color pattern for each of the robots 3A to 3C disposed in the other robots 3A to 3C. As described above, the present invention is not limited to this, and various other imaging means can be applied.

  Furthermore, in the first embodiment described above, the color pattern detection means for detecting the color pattern of the identification body 23 imaged by the camera 16 based on the image information (image signal S1) supplied from the camera 16 is shown in FIG. Although the case where the plurality of color extraction units 31A to 31U and the color pattern detection unit 32 configured as described above are configured has been described, the present invention is not limited thereto, and various other configurations can be applied.

  Further, in the first embodiment described above, based on the color pattern of the identification body 23 detected by the color pattern detection unit 32 and the color pattern information of the identification body 23 for each of the robots 3A to 3C stored in advance. Although the identification means for identifying the individual robots 3A to 3C having the identification body 23 imaged by the camera 16 is configured by the comparison / calculation unit 33 and the first memory 34, the present invention has been described. Not limited to this, various other configurations can be applied. In this case, various other storage means can be applied in place of the first memory 34.

  Further, in the first embodiment described above, a plurality of size detecting means for detecting the diameter (or the size of other portions) of the identification body 23 in the image based on the image signal S1 supplied from the camera 16 are provided. Although the case where the color extraction units 31A to 31U and the color pattern detection unit 32 are configured has been described, the present invention is not limited to this, and various other configurations can be applied.

  Further, in the first embodiment described above, the distance L1 from the camera 16 to the identifier 23 is calculated based on the size of the identifier 23 detected by the color pattern detector 32 and the reference value stored in advance. In the above description, the calculation unit is configured by the comparison / calculation unit 33 and the second memory 35. However, the present invention is not limited to this, and various other configurations can be applied. In this case, various other storage means can be applied in place of the second memory 35.

  Further, in the first embodiment described above, the reference value for calculating the distance L1 from the camera 16 to the identification body 23 is the pixel of the identification body 23 when the camera 16 and the identification body 23 are separated by 1 [m]. Although the case where the diameter is selected as a unit has been described, the present invention is not limited to this, and various other values can be applied.

  Furthermore, in the first embodiment described above, the case where each identifier 23 is simply colored in a unique color pattern with a plurality of colors has been described. However, the present invention is not limited to this, for example, the color of each identifier 23 Each pattern is represented in a predetermined color space (for example, the HSI space described with reference to FIG. 16, the RGB space in which the colors are represented by red, green, and blue levels, or the color is represented by the luminance level and the first and second color difference levels. Or a combination of a plurality of colors separated by a predetermined first distance (that is, a color separated by a first distance in a predetermined color space is used as a color for identification). good.

  In this case, instead of the color extraction units 31A to 31U, the individual identification unit 30 of each of the robots 3A to 3C converts the image signal S1 supplied from the camera 16 into an image signal having an image format corresponding to the color space. Means (for example, the image format conversion unit 93 of the fourth embodiment) and a plurality of color extraction means (for example, pixels for extracting different designated color pixels from the image based on the image signal output from the conversion means) Each color extraction unit 94A to 94D) of the fourth embodiment is provided, and the image is picked up by the camera 16 by a color pattern detection unit composed of the conversion unit, the plurality of color extraction units, and the color pattern detection unit 32 (FIG. 4). The color pattern of the identified body 23 may be detected. By doing so, it is possible to avoid an erroneous determination of the color colored on the identification body 23 in the robots 3A to 3C, and to identify an individual identification system and a robot that can identify the individual of the robots 3A to 3C with higher accuracy. realizable. In this case, if the image format of the image signal S1 output from the camera 16 is in accordance with the color space described above, the conversion means can be omitted.

  Furthermore, in this case, as a combination of colors applied to the identification body 23, adjacent colors may be selected so as to be separated by a second distance larger than the first distance in the above-described color space. By doing so, it is possible to realize an individual identification system and a robot that can identify individual robots 3A to 3C with higher accuracy.

  Furthermore, in the above-described first embodiment, the case where an object that does not emit light is applied as the identifier 23 is described. However, the present invention is not limited to this, and the identifier 23 may be, for example, a transparent spherical shell. While the light emitting means such as a light bulb is housed, the surface of the spherical shell may be colored with a corresponding color pattern so that the identification body 23 emits light of the corresponding color pattern. It is possible to reduce the probability of erroneous identification occurring in color identification by the identification means, and to realize an individual identification system and a robot that can identify individual robots 3A to 3C with higher accuracy.

  Further, in the second embodiment described above, the case where the color pattern is formed by color-coding the surface of the identification seal 52 as shown in FIG. 10 is described, but the present invention is not limited to this, and the identification seal is not limited thereto. The surface of 52 may be color-coded as shown in FIG. 32. In short, if the surface of the identification seal 52 is color-coded by a plurality of colors, various other shapes are applied as the shape of the color pattern. it can.

  In this case, for example, as shown in FIG. 33, a color pattern may be formed by coloring a plurality of concentric annular regions 141A to 141C using a predetermined number of colors from among a plurality of types of identification colors. good.

  In addition to this, for example, as shown in FIGS. 34 (A) to (C), linear regions 141D colored in a predetermined color extending in a predetermined radial direction of the annular regions 141A to 141C, or rectangular regions not colored. A region such as 141E or a fan-shaped region 141F may be provided. In this case, by setting in advance such that a region having a predetermined shape such as a straight region 141D, a rectangular region 141E, or a fan-shaped region 141F faces the front direction or the rear direction of the robots 51A to 51C, the robot is set based on the direction of the region. The directions of 51A to 51C can also be detected.

  Further, in the second embodiment described above, the case has been described in which the identification stickers 52 whose surfaces are color-coded with a predetermined color pattern are attached as means for applying different color patterns to the robots 51A to 51C. The present invention is not limited to this. For example, the robots 51A to 51C are given different predetermined color patterns by directly painting the upper surfaces of the robots 51A to 51C (or other predetermined positions that can be imaged by the camera 53). Alternatively, various other methods can be applied as means for giving different predetermined color patterns to the robots 51A to 51C.

  In this case, for example, light emitting means for emitting light of different color patterns to each of the robots 51A to 51C (for example, a light emitter such as a light bulb is disposed below the transparent film and colored with a color pattern corresponding to the transparent film). In this way, it is possible to reduce the probability of erroneous identification occurring in the color identification by the identification means, and to identify the individual robots 3A to 3C more accurately. An individual identification system and a robot can be realized.

  Further, in the second embodiment described above, the robots 51A to 51A in the region 2 are based on the positions of the color patterns of the robots 51A to 51C in the image based on the image information (image signal S1) supplied from the camera 53. Although the case where the position detecting means for detecting the position of 51C is configured by the comparison / calculation unit 56 and the memory 57 has been described, the present invention is not limited to this, and various other configurations can be applied. In this case, various other storage means can be applied in place of the memory 57.

  Further, in the second embodiment described above, a case has been described in which the positions of the robots 51A to 51C detected by the individual identification unit 54 are transmitted to the robots 51A to 51C by radio waves. However, the present invention is not limited to this. However, other wireless means such as infrared rays or various other transmission means such as wire can be applied.

  Further, in the second embodiment described above, the case where each identification seal 52 is simply color-coded into a unique color pattern with a plurality of colors has been described. However, the present invention is not limited to this, for example, the color of each identification seal 52 Each pattern is formed by a combination of a plurality of colors separated by a predetermined first distance in a predetermined color space (for example, HSI space, RGB space, YUV space, etc.) (that is, in the predetermined color space as a color for identification) It is also possible to use a color separated by the first distance.

  In this case, the individual identification unit 54 replaces the color extraction units 31A to 31U with conversion means (for example, a fourth embodiment) that converts the image signal S30 supplied from the camera 53 into an image signal having an image format corresponding to the color space. The image format conversion unit 93) of the example and a plurality of color extraction means (for example, each color of the fourth embodiment) for extracting pixels of specified colors different from each other based on the image signal output from the conversion means Each of the identification stickers 52 imaged by the camera 53 by the color pattern detection means constituted by the conversion means, the plurality of color extraction means, and the color pattern detection part 55 (FIG. 11). A color pattern may be detected. By doing so, it is possible to avoid an erroneous determination of the color colored on the identification seal 52 in the individual identification unit 54, and to realize an individual identification system that can identify the individual robots 51A to 51C with higher accuracy. . In this case as well, the conversion means can be omitted if the image format of the image signal S1 output from the camera 16 is in accordance with the color space described above.

  Further, in this case, as a combination of colors applied to the identification seal 52, adjacent colors may be selected so as to be separated by a second distance larger than the first distance in the color space described above. By doing so, it is possible to realize an individual identification system capable of identifying the individual robots 51A to 51C with higher accuracy.

  Furthermore, in the above-described third embodiment, the case where the color to be applied is changed for each of the wall surfaces 72AA to 72AD along each side of the region 2 has been described. The wall surfaces 72AA to 72AD along each side may be divided into a plurality of regions parallel to the Z direction, and the color to be applied may be changed for each region.

  Furthermore, in the first to third embodiments described above, the case where the color extraction units 31A to 31U are configured as shown in FIG. 6 has been described, but the present invention is not limited to this, and the main point is to be supplied. Separating means for separating an image signal into a luminance signal and a color difference signal (in the embodiment, based on the luminance signal and the color difference signal, the luminance level and the color difference level are sequentially detected for each pixel in the image based on the luminance signal and the color difference signal. Level detection means (analog / digital conversion circuits 42, 43A, 43B in the embodiment), the luminance level and color difference level of the pixel detected by the level detection means, and the upper limit value of the color difference level for each brightness level stored in advance And determining means for determining whether or not each pixel has a predetermined color based on the lower limit value (in the embodiment, the first to fourth memories 44A to 44D and the first to fourth comparison circuits 45A to 45A). If 5D and judging circuit 48) and of being so provided, the configuration of the color extraction unit, can be applied to various other configurations.

  Furthermore, in the above-described third to eighth embodiments, among predetermined wall surfaces 72AA to 72AD, 131AA to 131AD of different colors provided along the periphery of the region 2, corresponding predetermined wall surfaces 72AA to 72AD, 131AA are provided. ~ 131AD (that is, the front wall surfaces 72AA to 72AD, 131AA to 131AD) has been described as the case where the camera 16 is applied. However, the present invention is not limited to this, and various other imaging means are used. Applicable.

  Further, in the third to eighth embodiments described above, based on the image information (image signal S1) output from the camera 16, the colors of the wall surfaces 72AA to 72AD and 131AA to 131AD imaged by the camera 16 and the wall surfaces thereof. 72AA-72AD, 131AA-131AD relative position (the distance L2 from the wall surface 72AA-72AD, 131AA-131AD or one end of the wall surface 72AA-72AD, 131AA-131AD in the direction parallel to the wall surface 72AA-72AD, 131AA-131AD The color and relative position detection means for detecting the distance x) from a plurality of color extraction units 31A to 31D, 94A to 94E, wall detection units 8, 103, 114, 123, comparison / calculation units 82, 104, 115, 124 and first and second memories 83, 84, 105, 1 It has dealt with the case of constituting the 6,116,117,125,126, the present invention is not limited thereto and can be applied to various other configurations. In this case, various other storage means can be applied instead of the first and second memories 83, 84, 105, 106, 116, 117, 125, 126.

  Furthermore, in the above-described third to eighth embodiments, the colors of the wall surfaces 72AA to 72AD and 131AA to 131AD imaged by the camera 16 and the relative positions with respect to the wall surfaces 72AA to 72AD and 131AA to 131AD, and all of the previously stored information. Position detecting means for detecting the position in the region 2 based on the color of each of the wall surfaces 72AA to 72AD, 131AA to 131AD and the map information of the region 2 including the positions of all the wall surfaces 72AA to 72AD, 131AA to 131AD The comparison / calculation units 82, 104, 115, 124 and the third memories 85, 107, 118, 127 are described. However, the present invention is not limited to this, and the third memory 85, Various other storage means can be applied in place of 107, 118 and 127.

  Further, in the third to eighth embodiments described above, the robots 71, 91, 101, 111, 121 detect their positions in the region 2 from the two wall surfaces 72AA-72AD, 131AA-131AD. As described above, the present invention is not limited to this, and the position of the user in the region 2 may be detected from the three or more wall surfaces 72AA to 72AD and 131AA to 131AD.

  Further, in the above-described fourth to eighth embodiments, as a color for identification, a color having an angle around the origin O in the polar coordinates shown in FIG. Although the case of doing so has been described, the present invention is not limited to this, and a color in which the angle around the origin O in the polar coordinates shown in FIG. Further, as a color for identification, a color separated by a predetermined distance in a color space other than the HSI space (RGB space, YUV space, etc.) may be used.

  Further, in the above-described fourth to eighth embodiments, a color having an angle around the origin O in the polar coordinates shown in FIG. 16B in the HSI space that is separated by 60 [°] or more is simply selected as a color for identification. However, the present invention is not limited to this. For example, a color separated by 60 [°] or more in the HSI space (or other color space) is assigned to adjacent wall surfaces (for example, FIG. 16B). The wall surface 72AA may be yellow (Y), the wall surface 72AB may be cyan (C), the wall surface 72AC may be red (R), and the wall surface 72AD may be coated with blue (B). By doing so, it is possible to avoid erroneous determination when determining the colors of the wall surfaces 72AA to 72AD131AA to 131AD inside the robots 91, 101, 111, and 121, and the position of the user in the region 2 with higher accuracy. It is possible to construct a position detection system and a robot apparatus that can detect the above.

  Further, in the fifth embodiment described above, the case where the saturation (saturation) of the color applied to each of the wall surfaces 72AA to 72AD is linearly changed has been described. However, the present invention is not limited to this and is nonlinear. It may be changed. However, in this case, it is necessary to prevent portions having the same saturation from being present on one of the wall surfaces 72AA to 72AD.

Furthermore, in the above-described fifth embodiment, as predetermined data relating to the change in the degree of saturation of each wall surface 72AA to 72AD, the length M of each wall surface 72AA to 72AD and the saturation degree at one end and the other end of each wall surface 72AA to 72AD. Although the case where S _min and S _max are stored in the second memory 106 has been described, the present invention is not limited to this. In short, the data and the image signal S50 detected by the wall detecting unit 103 are used. If comparison and calculation section 104 on the basis of the saturation S _x color applied to the wall surface 72AA~72AD at the center of the image based on that the data can detect its position in the area 2 For example, data other than this can be applied as data stored in the second memory 106.

  Furthermore, in the sixth embodiment described above, the case where the slanted lines 112 are indicated on the wall surfaces 72AA to 72AD from the vicinity of the lower end of one end of the wall surfaces 72AA to 72AD to the vicinity of the upper end of the other end has been described. The invention is not limited to this. For example, as shown in FIG. 35, diagonal lines 112 may be written on the wall surfaces 72AA to 72AD so as to reach the upper end of the other end from the lower end of one end of the wall surfaces 72AA to 72AD. A slanted line may be written in other forms.

Furthermore, in the above-described sixth embodiment, as predetermined data regarding the change in the ratio of the portion above the oblique line 112 to the height of the wall surfaces 72AA to 72AD in each of the wall surfaces 72AA to 72AD, the length M of each wall surface 72AA to 72AD, A case will be described in which the lengths U _a1 , U _b1 , L _a1 , and L _{b1 of the} upper and lower portions of the wall surfaces 72AA to 72AD at one end and the other end of the wall surfaces 72AA to 72AD are stored in the second memory 117. However, the present invention is not limited to this, and the point is that the data is higher than the oblique line 112 with respect to the height of the wall surfaces 72AA to 72AD in the central portion of the image based on the image signal S50 detected by the wall detection unit 115. can comparison and calculation section 115 on the basis of the ratio R _x part (or the lower part) to detect its own position within the region 2 If you are a chromatography data, as the data stored in the second memory 117 can be applied to other data.

  Further, in the seventh embodiment described above, the virtual line K1 that separates the upper and lower portions of the wall surfaces 72AA to 72AD passes through the vicinity of the lower end of one end of the wall surfaces 72AA to 72AD and the vicinity of the upper end of the other end. As described above, the present invention is not limited to this. For example, as shown in FIG. 36, each of the virtual lines K1 (or may be diagonal lines) passes through the lower end of one end of the wall surfaces 72AA to 72AD and the upper end of the other end. The wall surfaces 72AA to 72AD may be divided into an upper portion and a lower portion, and a virtual line K1 or a diagonal line may be selected or indicated in other forms.

Further, in the seventh embodiment described above has dealt with the case where the ratio R _x of length H _x of the high saturation portions with respect to the height of the wall 72AA~72AD was made to vary linearly, the present invention is not limited thereto Alternatively, it may be changed nonlinearly. However, in this case, it is necessary that the ratio _{R x} of length _{H x} of the high saturation portions with respect to the height of the wall 72AA~72AD in one wall 72AA~72AD there is no portion to be the same.

Further, in the seventh embodiment described above, the predetermined data relating to the change in the ratio _{R x} of length _{H x} of the high saturation portions with respect to the height of the wall 72AA～72AD in each wall 72AA～72AD, the length of each wall 72AA～72AD M and the lengths H _a2 , L _a2 , H _b2 , and L _b2 of the high saturation portion and the low saturation portion at one end and the other end of each of the wall surfaces 72AA to 72AD are stored in the second memory 126. Although the present invention is not limited to this, the point is that the length of the high saturation portion with respect to the height of the wall surfaces 72AA to 72AD in the center portion of the image based on the data and the image signal S50 detected by the wall detection unit 123. if it's data which can detect their position in the H _x ratio _of R _x and comparison and calculation section 124 in region 2 on the basis of, the As data to be stored in the memory 126 can be applied to other data.

  Further, in the third and eighth embodiments described above, distances from the wall surfaces 72AA to 72AD and 131AA to 131AD are detected as relative positions to the wall surfaces 72AA to 72AD and 131AA to 131AD in the robot 91, and the fourth to fourth embodiments are detected. In the eighth embodiment, the distances from one end of the wall surfaces 72AA to 72AD in the direction parallel to the wall surfaces 72AA to 72AD are detected as relative positions to the wall surfaces 72AA to 72AD in the robots 101, 111, and 121. Although the case has been described, the present invention is not limited to this, and other relative positions may be detected.

  Furthermore, in the above-described eighth embodiment, the case where each of the wall surfaces 131AA to 131AD is configured by the panel surface of the liquid crystal panel display 132 has been described. However, the present invention is not limited to this, and each of the wall surfaces 131AA to 131AD is a liquid crystal. You may make it comprise with the display surface of displays (for example, CRT (Cathode Ray Tube) etc.) other than the panel display 132. FIG. Further, for example, a normal wall surface 72AA to 72AD (FIG. 13) is provided with a recess, a film of a predetermined color is attached so as to cover the recess, and a light source is disposed in the recess so that the wall surface 72AA to 72AD is provided. The light of the assigned color may be emitted toward the region 2. In short, the wall surface is constituted by the light emitting means for emitting the light 2 of the color assigned to the wall surface toward the region, or the light is emitted. As long as the means is provided on the wall surface, various other configurations can be applied as the configuration of the light emitting means.

  Furthermore, in the above-described eighth embodiment, the case where each liquid crystal panel display 132 simply emits light of a different color for each of the wall surfaces 131AA to 131AD has been described, but the present invention is not limited to this, for example, each The assigned colors and oblique lines may be displayed on the liquid crystal panel display 132 as in the fifth to seventh embodiments, and the robot 91 may be configured as in the fifth to seventh embodiments.

  Further, in the fourth to seventh embodiments described above, the case where the image format conversion unit 93 is provided to convert the image format of the image signal S1 output from the camera 16 into an image format corresponding to the HSI space will be described. However, the present invention is not limited to this. For example, when the image format of the image signal S1 output from the camera 16 is an image format corresponding to the color space used when selecting the colors of the wall surfaces 72AA to 72AD. The image format conversion unit 93 may be omitted.

  The present invention can be widely applied to various robot devices and the like in addition to autonomous mobile robots.

It is a top view which shows the whole structure of the identification system by 1st Example. It is a rough-line side view which shows the structure of the robot by 1st Example, and the structure of an identification body. 1 is a schematic block diagram illustrating a configuration of a robot according to a first embodiment. It is a block diagram which shows the structure of the individual identification part by 1st Example. It is a basic diagram with which it uses for description of a structure of a color extraction part. It is a block diagram which shows the structure of a color extraction part. It is a top view which shows the whole structure of the identification system by 2nd Example. It is a top view which shows the whole structure of the identification system by 2nd Example. It is a rough-line perspective view which shows the structure of the robot by 2nd Example. It is a top view with which it uses for description of an identification seal | sticker. It is a block diagram which shows the structure of the individual identification part by 2nd Example. It is a rough-line block diagram which shows the structure of the robot by 2nd Example. It is a perspective view which shows the whole structure of the position identification system by 3rd Example. It is a top view which shows the whole structure of the position identification system by 3rd Example. It is a block diagram which shows the structure of the position detection part by 3rd Example. It is an approximate line figure used for explanation of HSI space. It is a top view which shows the whole structure of the position identification system by 4th Example. It is a block diagram which shows the structure of the position detection part by 4th Example. It is a top view which shows the whole structure of the position identification system by 5th Example. It is a basic diagram which shows the mode of each wall surface in 5th Example. It is a block diagram which shows the structure of the position detection part by 5th Example. It is a top view which shows the whole structure of the position identification system by 6th Example. It is a basic diagram which shows the mode of each wall surface in 6th Example. It is a block diagram which shows the structure of the position detection part by 6th Example. It is a top view which shows the whole structure of the position identification system by 7th Example. It is a basic diagram which shows the mode of each wall surface in 7th Example. It is a block diagram which shows the structure of the position detection part by 7th Example. It is a top view which shows the whole structure of the position identification system by 8th Example. It is a top view with which it uses for description of the structure of the position identification system by 8th Example. It is a rough-line side view which shows another Example. It is the side view and perspective view which show another Example. It is a top view which shows another Example. It is a top view which shows another Example. It is a top view which shows another Example. It is a top view which shows another Example. It is a top view which shows another Example.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1,50 ... Individual identification system, 2 ... Area, 3A-3C, 51A-51C, 71, 91, 101, 111, 121 ... Robot, 16, 53 ... Camera, 23 ... Identification object, 30, 54... Individual identification unit, 31 A to 31 U, 94 A to 94 E... Color extraction unit, 32 and 55... Color pattern extraction unit, 33, 56, 82, 104, 115, 124. 35, 44A-44D, 57, 83-85, 105-107, 116-118, 125-127 ... Memory, 36 ... CPU, 41 ... Separation circuit, 42, 43A, 43B ... Analog / digital conversion circuit 45A to 45D... Comparison circuit 48. Determination circuit 49... Frame memory 52 52 Identification seal 58... Transmitting unit 70 70 Wall 72 AA to 72 AD 131 AA to 131 A ...... Wall surface, 80, 102, 113, 122 …… Position detector, 81, 103, 114, 123 …… Wall detector, 93 …… Image format converter, 132 …… Liquid crystal panel, 133 …… Control unit, D1: Luminance data, D2A, D2B ... Color difference data, S1, S30, S50 ... Image signal, S10A to S10U, S51A to S51D ... Color extraction signal, S11 ... Discriminator information signal, S12 ... Discriminator Detection signal, S20 ... Luminance signal, S21A, S21B ... Color difference signal, S31 ... Identification sticker detection signal, S32 ... Robot position detection signal, S40 ... Wall detection signal, S41, S62, S71, S82 ... Position Detection signal.

Claims (29)

In an identification apparatus for identifying a moving body that moves within a predetermined area,
An identifier of a different color pattern provided on each of the moving bodies;
An imaging unit that is provided in each of the moving bodies and that images the identification bodies disposed in the other moving bodies;
Color pattern detection means for detecting the color pattern of the identifier imaged by the imaging means based on the first image information supplied from the imaging means;
Based on the detection result by the color pattern detection means and the color pattern information of the identification body for each of the mobile bodies stored in advance, the mobile body having the identification body imaged by the imaging means is identified. An identification device comprising: identification means. A size detecting means for detecting the size of the identifier in the image based on the first image information supplied from the imaging means;
The identification device according to claim 1, further comprising a calculation unit that calculates a distance to the identification object based on a detection result by the detection unit and a reference value stored in advance. The identifier is
It is a rotating body having a central axis perpendicular to the moving direction of the moving body, and its surface is divided into a plurality of band-like areas parallel to the moving direction of the moving body, and each of the band-like areas has a predetermined color pattern. The identification device according to claim 1, wherein the identification device is colored in a predetermined color. Each of the color patterns is formed by a combination of a plurality of colors separated from each other by a predetermined first distance in a predetermined color space,
The color pattern detection unit converts the first image information supplied from the imaging unit into second image information corresponding to the corresponding color space, and then performs the imaging based on the second image information. The identification device according to claim 1, wherein the color pattern of the identification object imaged by means is detected. 5. Each of the color patterns is formed by selecting the color combination so that adjacent colors are separated from each other by a second distance larger than the first distance in the color space. The identification device described in 1. The identification device according to claim 1, wherein each identification body emits light of the corresponding color pattern. In an identification apparatus for identifying a moving body that moves within a predetermined area,
An imaging unit that is provided in each of the moving bodies, and that captures an identification body having a different color pattern for each of the moving bodies provided in the other moving bodies;
Color pattern detection means for detecting the color pattern of the identifier imaged by the imaging means based on the first image information supplied from the imaging means;
Based on the detection result by the color pattern detection means and the color pattern information of the identification body for each of the mobile bodies stored in advance, the mobile body having the identification body imaged by the imaging means is identified. An identification device comprising: identification means. A size detecting means for detecting the size of the identifier in the image based on the first image information supplied from the imaging means;
The identification device according to claim 7, further comprising: a calculation unit that calculates a distance to the identification object based on a detection result by the detection unit and a reference value stored in advance. The identifier is
It is a rotating body having a central axis perpendicular to the moving direction of the moving body, and its surface is divided into a plurality of band-like areas parallel to the moving direction of the moving body, and each of the band-like areas has a predetermined color pattern. The identification device according to claim 7, wherein the identification device is colored in a predetermined color. Each of the color patterns is formed by a combination of a plurality of colors separated from each other by a predetermined first distance in a predetermined color space,
The color pattern detection unit converts the first image information supplied from the imaging unit into second image information corresponding to the corresponding color space, and then performs the imaging based on the second image information. The identification device according to claim 7, wherein the color pattern of the identification object imaged by means is detected. In an identification method for identifying a moving object that moves within a predetermined area,
A first step of providing each moving body with a different color pattern identification body;
A second step of imaging the identification body of the other mobile body by imaging means provided in each of the mobile bodies;
A third step of detecting the color pattern of the identifier imaged by the imaging means based on the first image information output from the imaging means;
And a fourth step of identifying the identification object imaged by the imaging means based on the detected color pattern of the identification object and color pattern information of each of the identification objects stored in advance. An identification method characterized by. A detection step of detecting a size of the identifier in an image based on the first image information output from the imaging means;
The identification method according to claim 11, further comprising a calculation step of calculating a distance to the identification object based on the detected size of the identification object and a reference value stored in advance. The identifier is
It is a rotating body having a central axis perpendicular to the moving direction of the moving body, and its surface is divided into a plurality of band-like areas parallel to the moving direction of the moving body, and each of the band-like areas has a predetermined color pattern. The identification method according to claim 11, wherein the identification method is colored in a predetermined color. Each of the color patterns is formed by a combination of a plurality of colors separated by a predetermined first distance in a predetermined color space,
In the third step, the first image information supplied from the imaging unit is converted into second image information corresponding to the corresponding color space, and then the imaging is performed based on the second image information. The identification method according to claim 11, wherein the color pattern of the identification object imaged by a means is detected. Each of the color patterns is formed by selecting a combination of the colors so that adjacent colors are separated from each other by a second distance larger than the first distance in the color space. 14. The identification method according to 14. The identification method according to claim 11, wherein each identification body emits light of the corresponding color pattern. In an identification apparatus for identifying a plurality of moving objects that move within a predetermined area,
Imaging means for imaging the entire region;
Color pattern detection means for detecting a different color pattern for each of the moving bodies respectively given to a predetermined position of each of the moving bodies based on the first image information output from the imaging means;
An identification means for identifying each moving body based on a detection result by the color pattern detecting means and information on the color pattern given to each moving body stored in advance. Identification device. Position detecting means for detecting the position of each moving body in the region based on the position of the color pattern of each moving body in the image based on the first image information supplied from the imaging means. The identification device according to claim 17 , wherein: 18. The identification device according to claim 17 , wherein each of the color patterns is formed by coloring a plurality of concentric annular regions using a predetermined number of colors from among a plurality of types of predetermined colors. The identification device according to claim 17 , wherein each of the color patterns includes a region extending in a radial direction of the annular region. Each of the color patterns is formed by a combination of a plurality of colors separated by a predetermined first distance in a predetermined color space,
The color pattern detection unit converts the first image information supplied from the imaging unit into second image information corresponding to the corresponding color space, and then converts each of the above-described color image detection units based on the second image information. The identification device according to claim 17 , wherein each of the color patterns of the moving body is detected. Each of the color patterns is formed by selecting a combination of the colors so that the adjacent colors are separated from each other by a second distance larger than the first distance in the color space. Item 22. The identification device according to Item 21 . The identification device according to claim 17 , further comprising: a light emitting unit that is provided in each of the moving bodies and emits light of the corresponding color pattern. In an identification method for identifying a plurality of moving objects that move within a predetermined area,
A first step of arranging imaging means for imaging the entire region at a predetermined position and applying different color patterns to the predetermined positions of each of the moving bodies;
A second step of detecting each of the color patterns of the moving bodies based on first image information output from the imaging means;
And a third step of identifying each moving body based on the detection result in the second step and the color pattern provided for each moving body stored in advance. How to identify. A detecting step of detecting the position of each moving body in the region based on the position of the color pattern of each moving body in the image based on the first image information supplied from the imaging means; The identification method according to claim 24 , wherein: Each of the above color patterns
25. The identification method according to claim 24 , wherein each of the plurality of concentric annular regions is colored using a predetermined number of colors from among a plurality of predetermined colors. Each of the above color patterns
The identification method according to claim 24 , further comprising a region extending in a radial direction of each annular region. Each of the color patterns is formed by a combination of a plurality of colors separated by a predetermined first distance in a predetermined color space,
In the second step,
After the first image information supplied from the imaging means is converted into second image information corresponding to the corresponding color space, the color pattern of each moving body is converted based on the second image information. 25. The identification method according to claim 24 , wherein each is detected. Each of the color patterns is formed by selecting a combination of the colors so that the adjacent colors are separated from each other by a second distance larger than the first distance in the color space. Item 30. The identification device according to Item 28 .

Images