JP5084756B2 - Autonomous mobile wheelchair - Google Patents

Description

  The present invention relates to a wheelchair that is used by an elderly person, a disabled person, and the like, and that can move by autonomously determining a moving direction and a moving speed while the user is on the wheelchair.

Patent Document 1 below discloses a robot that moves following a human. This robot is equipped with a camera, and recognizes the position of the tracking target person and an obstacle appearing in the moving direction from the image of the camera, and determines the moving direction and moving speed.
The inventor of the present invention first identifies not only the position of the person to be followed (caregiver) and the obstacle in the running direction but also the direction of the caregiver's face and the direction in which the caregiver is gazing in the wheelchair that moves autonomously. And the very intelligent wheelchair which autonomously controls a moving direction and a moving speed is proposed (the following patent document 2).

JP 2008-234400 A Japanese Patent Application No. 2008-27708

In Japan, where the birthrate is declining and the population is aging, the number of people using wheelchairs is expected to continue to increase. Correspondingly, in order to keep wheelchair users engaged in society, for example, facilities such as museums and libraries can be used in the same way as normal people, or participate in various events without hesitation. It is necessary to prepare an environment where
On the other hand, the declining birthrate and aging population are reducing the number of wheelchair users on the side of care and support.
For this reason, the wheelchair is a place where people can go and go with healthy people, and it can move smoothly without burdening caregivers and without disturbing other people. A good mechanism and structure are required.

  The present invention was devised in view of such circumstances, and an object of the present invention is to provide an autonomous mobile wheelchair capable of intelligent autonomous behavior according to the surrounding situation and capable of downsizing the sensor mechanism. Yes.

The present invention is a wheelchair that moves autonomously according to a caregiver's operation, and includes sensor means that provides information representing the surrounding situation, and caregiver that detects the position of the caregiver based on the information of the sensor means. A caregiver based on information on the distance and direction from the sensor means to the caregiver provided from the sensor means , an ambient condition detection means for detecting surrounding people and obstacles based on the information of the sensor means, A carer's torso detecting means for detecting a contour shape obtained by cutting the torso of the body, and a moving path for determining a moving path of the wheelchair based on the detection results of the caregiver position detecting means , the surrounding state detecting means and the caregiver torso detecting means Discriminating means , wherein the caregiver torso detecting means discriminates the orientation of the caregiver's body from the detected substantially elliptical contour shape, and the movement path discriminating means is detected by the caregiver torso detecting means. Is In consideration of information about the caregiver orientation and determines the movement path.
This autonomous mobile wheelchair can determine the movement route according to the surrounding situation, can know the caregiver's orientation from the contour shape of the carer's torso, and know the caregiver's orientation, An appropriate route can be determined even when the user is next to the person.

In the autonomous mobile wheelchair according to the present invention, the sensor means includes a combination of an omnidirectional camera and a distance sensor, and corresponds to information on the distance and direction to the subject with respect to the subject reflected in the omnidirectional camera. The attached image information is provided.
This autonomous mobile wheelchair can determine the travel route of the wheelchair based on information from a compact sensor that combines an omnidirectional camera and a distance sensor.

Further, in the autonomous mobile wheelchair according to the present invention, the movement route determination means can determine the movement route while interlacing the route following the caregiver and the route guiding the caregiver.
Depending on the surrounding situation, "following" that selects the route so as not to leave the caregiver's position, and the wheelchair itself selects the optimal route to the destination, and the caregiver follows the "guidance" Can be switched automatically.

The autonomous mobile wheelchair according to the present invention further includes a caregiver face direction detecting unit that detects a face direction of the caregiver based on information of the sensor unit, and the moving path determination unit includes the caregiver face. The moving path can be determined in consideration of the detection result of the direction detecting means.
By knowing the direction of the caregiver's face, the direction of the caregiver's progress can be estimated with higher accuracy, and by detecting the object in the direction of interest from the direction of the caregiver's face, it can be used as a landmark The moving direction can be determined.

The autonomous mobile wheelchair according to the present invention further includes a passenger face direction detecting unit that detects a face direction of the passenger based on information of the sensor unit, and the moving path determination unit includes the passenger face. The moving path can be determined in consideration of the detection result of the direction detecting means.
By detecting the object in the direction of interest from the direction of the passenger's face, the moving direction can be determined using this as a landmark of the destination. When the caregiver is lost, the caregiver can be redetected by searching for the direction of the face of the passenger.

The autonomous mobile wheelchair according to the present invention further includes a front face detection means for detecting a front face facing the direction of the wheelchair based on information of the sensor means, and the movement path determination means includes the front face detection. The movement route can be determined in consideration of the detection result of the means.
In situations where the caregiver is lost and temporarily separated from the caregiver, the person looking at the wheelchair is often a caregiver, so the caregiver can be re-detected by looking for the front face. it can.

The autonomous mobile wheelchair according to the present invention further includes a zoom camera, and the caregiver face direction detection means, the passenger face direction detection means or the front face detection means uses the image of the zoom camera to perform the care. The face direction of the passenger, the face direction of the passenger, or the front face may be confirmed.
The direction in which the zoom camera is directed to capture the face image can be accurately estimated from the information of the sensor means. For example, the situation where the caregiver and the passenger are conversing with each other can be detected from the face image of the zoom camera. In that case, the state where the caregiver is attached is maintained and the conversation is easy. can do.

The autonomous mobile wheelchair according to the present invention further includes a moving body tracking unit that tracks the movement of the moving body, and the moving path determination unit determines the moving path in consideration of the tracking result of the moving body tracking unit. can do.
By tracking the movement of a moving object and predicting the movement of the object based on the tracking history, it is possible to smoothly avoid a collision with the object.

The autonomous mobile wheelchair according to the present invention may further include information presenting means for presenting the traveling direction of the wheelchair determined by the travel route determining means.
By indicating the direction of travel of the wheelchair, surrounding people can act so as not to interfere with the progress of the wheelchair.

  The autonomous mobile wheelchair of the present invention is capable of intelligent autonomous behavior according to the surrounding situation using a compact sensor mechanism.

The figure which shows typically the autonomous mobile wheelchair in the 1st Embodiment of this invention. The figure which shows the sensor mechanism used by this invention The block diagram which shows the hardware constitutions of the autonomous mobile wheelchair in the 1st Embodiment of this invention. The block diagram which shows the function of the autonomous mobile wheelchair in the 1st Embodiment of this invention. Diagram showing laser image Diagram showing movement route selection pattern Flow chart showing movement route selection operation The block diagram which shows the function of the autonomous mobile wheelchair in the 2nd Embodiment of this invention. The figure explaining the detection operation | movement of a caregiver trunk detection means The figure which shows the movement route selection when the caregiver bends The block diagram which shows the function of the autonomous mobile wheelchair in the 3rd Embodiment of this invention. The figure which shows typically the autonomous mobile wheelchair in the 3rd Embodiment of this invention. The block diagram which shows the hardware constitutions of the autonomous mobile wheelchair in the 3rd Embodiment of this invention. The block diagram which shows the function of the autonomous mobile wheelchair in the 4th Embodiment of this invention. The block diagram which shows the function of the autonomous mobile wheelchair in the 5th Embodiment of this invention. The figure which shows typically the autonomous mobile wheelchair in the 5th Embodiment of this invention. The figure which shows typically the other example of the autonomous mobile wheelchair in the 5th Embodiment of this invention. The block diagram which shows the function of the autonomous mobile wheelchair in the 6th Embodiment of this invention. The figure which shows typically the autonomous mobile wheelchair in the 6th Embodiment of this invention. The block diagram which shows the hardware constitutions of the autonomous mobile wheelchair in the 6th Embodiment of this invention. The block diagram which shows the function of the autonomous mobile wheelchair in the 7th Embodiment of this invention The block diagram which shows the hardware constitutions of the autonomous mobile wheelchair in the 7th Embodiment of this invention. The figure which shows typically the autonomous mobile wheelchair in the 7th Embodiment of this invention.

(First embodiment)
FIG. 1 shows a first embodiment of the autonomous mobile wheelchair of the present invention. FIG. 1 shows a state in which an occupant 30 rides on the autonomous mobile wheelchair 10 and a caregiver 40 follows the side or behind it.
The autonomous mobile wheelchair 10 has a sensor mechanism 14 in which a laser side area sensor and an omnidirectional camera are combined with a support column standing at the rear of the wheelchair body.
As shown in FIG. 2, the sensor mechanism 14 irradiates a laser beam radially, measures a distance to a surrounding object by reflection from the object, and a cylindrical or hemispherical mirror. The omnidirectional camera 16 shoots an omnidirectional subject, and the omnidirectional camera 16 is mounted on the laser side area sensor 15 so as to be compact. By calibrating the positional relationship between the omnidirectional camera 16 and the laser side area sensor 15 in advance, the image information of the omnidirectional camera 16 and the distance information of the laser side area sensor 15 can be associated with each other. Provides information on the distance and direction to the subject as well as the omnidirectional subject image.
In addition, the autonomous mobile wheelchair 10 is equipped with a personal computer in the rear space.

FIG. 3 shows a hardware configuration of the autonomous mobile wheelchair 10. The drive system includes a wheel motor 13 that rotates wheels to move the wheelchair and change the direction of the wheelchair, and a motor control board 12 that controls the wheel motor 13. The sensor system includes an omnidirectional camera 16 and a laser side region. A CPU 17 that includes a sensor mechanism 14 including a sensor 15 and controls the autonomous mobile wheelchair 10 as a control system and performs arithmetic processing such as image information analysis processing, and a memory 11 that stores control programs and various data It has.
In addition, the autonomous mobile wheelchair 10 also has a joystick for driving operation, a power source, and the like, but these configurations not directly related to the present invention are omitted here.
The analysis of the image information obtained from the sensor mechanism 14 may be performed using a DSP (digital signal processor).

  As shown in FIG. 4, the autonomous mobile wheelchair 10 includes a caregiver position detection unit 22 that detects the position of the caregiver 40 based on detection information of the sensor mechanism 14, and a sensor mechanism. 14, a surrounding state detection unit 21 that detects surrounding people and obstacles based on the detection information 14, and a movement route that determines the movement route of the wheelchair based on the detection results of the caregiver position detection unit 22 and the surrounding state detection unit 21. The discriminating means 23 and movement control means 25 for moving the wheelchair along the determined movement path are provided.

  The caregiver position detection unit 22 is realized by processing of the CPU 17 that analyzes the information of the sensor mechanism 14 and detects the distance and direction to the caregiver 40. The caregiver position detection unit 22 detects the person who is closest to the information of the sensor mechanism 14 as the caregiver 40, and thereafter tracks the person as the caregiver 40.

  The surrounding state detection means 21 is realized by the processing of the CPU 17 that analyzes the information of the sensor mechanism 14 and detects the distance and direction of surrounding persons, obstacles, walls, and the like. As shown in FIG. 5, the surrounding state detection means 21 draws information on the distance to a person, an obstacle, and a wall obtained by a laser range sensor on a two-dimensional map (in the present invention, this is referred to as a “laser image”). It is possible to express the surrounding situation.

Moreover, the movement path | route discrimination | determination means 23 is implement | achieved by the process of CPU17 which selects the movement path | route of a wheelchair based on the detection result of the caregiver position detection means 22 and the surrounding condition detection means 21. FIG.
Further, the movement control means 25 is realized by the CPU 17 that performs motor control processing in accordance with a control program stored in the memory 11.

Next, the route selection process of the movement route discrimination means 23 will be described. FIG. 6 shows a route that the moving route determination means 23 selects when the caregiver 40 and the wheelchair 10 move in the direction of the person 60 who walks, as shown in FIG. FIG. 7 is a flowchart showing the processing procedure of the movement path determination means 23 at this time.
The movement route discrimination means 23 recognizes the position of the caregiver 40 from the information of the caregiver position detection means 22 and the surrounding situation detection means 21, and recognizes that the person 60 comes from the front, It is determined whether or not the ground is clear (step 1). When the destination is not clear, a place away from the caregiver in the traveling direction is set as the destination (step 2). A method for recognizing the destination will be described later.

When the destination is set, it is identified from the information of the surrounding situation detection means 21 whether there is a space where the caregiver and the wheelchair can enter on the caregiver side (step 3). It is determined whether the course of movement to the caregiver's side is significantly different from the direction of the destination (step 4). If not, the caregiver's side is avoided to avoid the person 60 as shown in FIG. (Step 5).
On the other hand, if the direction of the destination is significantly different in step 4, identify whether there is a space where the caregiver and the wheelchair can enter on the side of the wheelchair (step 6). It is determined whether or not the direction of travel to the destination is significantly different from the direction of the destination (step 7), and if not, as shown in FIG. 6 (c), move to the side of the wheelchair to avoid the person 60 (step 8). At this time, the wheelchair selects a route that the caregiver can easily follow and guides the caregiver. When the follow-up of the caregiver is slow or the distance is long, the speed is reduced or the course is corrected in a direction where the caregiver can join.

In step 7, if the direction of the destination is significantly different, if the caregiver and the wheelchair are separated, it is identified whether there is a space that can be entered (step 9). It is determined whether the direction is significantly different from the direction of the destination (step 10). If not, as shown in FIG. 6 (d), the person moves separately from the caregiver to avoid the person 60 (step 11).
If the direction of the destination is significantly different from the direction of the destination in step 10, the caregiver is followed as shown in FIG.
In this way, the movement route discrimination means 23 prompts the caregiver to follow the “following action” and the caregiver based on the information of the caregiver position detection means 22 and the surrounding situation detection means 21. Can be adaptively switched according to the surrounding situation.

“Following” is an operation of mainly selecting a route that does not leave the position of the caregiver. In this case, the route that the wheelchair itself is optimal in view of the surrounding situation is considered as a slave. On the other hand, “guidance” is an operation in which the wheelchair itself selects mainly the most appropriate route to go to the destination in consideration of the surrounding situation. In this case, the route that does not leave the caregiver's position is considered as a subordinate. Is done.
Therefore, “follow” and “guidance” are not uniquely determined by the positional relationship between the caregiver and the wheelchair. For example, even when the wheelchair is next to the caregiver, the wheelchair moves before the caregiver. In some cases, the caregiver may be guided, and the wheelchair may follow the caregiver's movement. There may also be cases where a wheelchair traveling slightly behind the caregiver guides.

(Second Embodiment)
As shown in FIG. 8, the autonomous mobile wheelchair according to the second embodiment includes the body of the caregiver 40 tracked by the caregiver position detection means 22 in addition to the configuration of the first embodiment (FIG. 4). A caregiver torso detecting means 24 for detecting the contour shape is provided.
The hardware configuration of the autonomous mobile wheelchair is the same as that of the first embodiment (FIG. 3), and the caregiver torso detection unit 24 analyzes the information of the sensor mechanism 14 to determine the contour shape of the carer's torso. This is realized by processing of the CPU 17 to detect.

  As shown in FIG. 9, the caregiver body detection unit 24 obtains a laser image obtained by horizontally scanning the body of the upright caregiver 40 from the information of the laser range sensor 15. The human torso has a substantially elliptical outline, but only a part of the elliptical outline can be obtained with a laser image. Therefore, for example, the caregiver torso detecting unit 24 performs pattern matching between the contour shape model of the person stored in the memory 11 in advance and the laser image, and specifies the direction of the ellipse. Then, the elliptical minor axis direction is identified as the front-rear direction of the body, and the major axis direction is identified as the lateral direction of the body.

  The relative position of the wheelchair relative to the caregiver can be accurately known from the information regarding the body orientation provided by the caregiver torso detection means 24 when the wheelchair is placed next to the caregiver. Even if there is no information on the caregiver torso detection means 24, it is possible to put it on the side of the caregiver, but by using the information on the caregiver torso detection means 24, the accuracy is increased, and it is next to the caregiver. When following or guiding in the attached state, an appropriate route can be determined.

For example, as illustrated in FIG. 10, the movement path determination unit 23 obtains information on the position of the caregiver 40 from the caregiver position detection unit 22 and moves the caregiver when the wheelchair 10 is moving side by side with the caregiver 40. Information on 40 body orientations is obtained from the caregiver torso detection means 24. And when caregiver 40 changes the direction of the body to the side of wheelchair 10 while progressing, it judges that he is going to turn in the direction of a wheelchair, and selects a path | route so that it may turn in the same direction as caregiver 40 (FIG. 10B).
By selecting such a route, it is possible to avoid a state where the wheelchair obstructs the course of the caregiver 40 who is about to turn (FIG. 10A).

At this time, if the wheelchair determines the route based only on the information on the position of the caregiver, the caregiver needs to quickly turn left to pass the wheelchair to avoid the collision in FIG. However, when using information on the body orientation, such a problem can be solved because the wheelchair can turn in the direction intended by the caregiver. The direction of the body of the caregiver 40 is reliable information for identifying the course of the caregiver 40.
Thus, the information on the body orientation of the caregiver 40 provided by the caregiver torso detection means 24 is extremely effective information when the wheelchair adaptively switches between “following” and “guidance”.

(Third embodiment)
As shown in FIG. 11, the autonomous mobile wheelchair of the third embodiment is directed to the face of the caregiver 40 being tracked by the caregiver position detection means 22 in addition to the configuration of the second embodiment (FIG. 8). Is provided with a caregiver face direction detecting means 26 for detecting the above.
The hardware configuration of this autonomous mobile wheelchair is the same as that of the first embodiment (FIG. 3), and the caregiver face direction detection means 26 analyzes the information of the sensor mechanism 14 and detects the face direction of the caregiver. This is realized by the processing of the CPU 17.
A method for detecting a face orientation from a face image has been widely known in the past, and is also described in detail in Patent Document 2.

When the caregiver face direction detection unit 26 detects the caregiver's face direction by a known method, the caregiver face direction detection unit 26 identifies whether the face direction is consistent with the body direction of the caregiver 40 obtained from the caregiver body detection unit 24. . By doing so, the caregiver's face orientation can be obtained with high accuracy.
Further, the movement route determination means 23 is detected by the caregiver face direction detection means 26 together with the position of the caregiver detected by the caregiver position detection means 22 and the orientation of the caregiver body detected by the caregiver body detection means 24. Taking into account the caregiver's face orientation, the traveling direction of the caregiver is estimated. Therefore, the traveling direction of the caregiver can be estimated with high accuracy.
Further, the movement route discrimination means 23 identifies an object that the caregiver is interested in from the caregiver face direction detected by the caregiver face orientation detection means 26 and the surrounding situation detected by the surrounding situation detection means 21. Select a travel route using as a landmark for the destination.

As shown in FIG. 12, a pan / tilt / zoom camera 50 may be separately provided for detecting the face direction of the caregiver. FIG. 13 shows a hardware configuration of the autonomous mobile wheelchair in this case.
The pan / tilt zoom camera 50 zooms in the direction of the caregiver's face detected by the caregiver's face orientation detection means 26 and projects the caregiver's face clearly. The caregiver face direction detection means 26 can detect the direction of the caregiver's face with high accuracy from the face image of the caregiver shown in the pan / tilt / zoom camera 50.

(Fourth embodiment)
As shown in FIG. 14, the autonomous mobile wheelchair according to the fourth embodiment detects a front face that faces the front of the face in the direction of the wheelchair in addition to the configuration of the third embodiment (FIG. 11). Face detecting means 27 is provided.
The hardware configuration of this autonomous mobile wheelchair is the same as that of the first embodiment (FIG. 3), and the front face detection means 27 performs image analysis of the information of the sensor mechanism 14 and the processing of the CPU 17 that detects the front face. Realized.
A method for detecting a front face from a face image is known and is described in detail in Patent Document 2.
When the wheelchair temporarily leaves the caregiver and heads for the destination, the caregiver may be hidden behind an obstacle such as a figure on the way, and the caregiver position detection means 22 may lose sight of the caregiver.
In such cases, the caregiver is likely facing the wheelchair. Therefore, the caregiver position detection means 22 that has lost sight of the caregiver considers the front face detected by the front face detection means 27 as a caregiver and resumes tracking.
Also in this case, as shown in FIG. 12, the pan / tilt / zoom camera 50 provided separately projects the front face detected by the front face detection means 27 clearly and confirms whether or not he / she is a caregiver. Anyway.

(Fifth embodiment)
As shown in FIG. 15, the autonomous mobile wheelchair of the fifth embodiment includes passenger face orientation detection means 28 that detects the face orientation of the passenger 30 in addition to the configuration of the fourth embodiment (FIG. 14). I have.
The hardware configuration of this autonomous mobile wheelchair is the same as that of the first embodiment (FIG. 3), but the sensor mechanism 14 is installed on the front side of the wheelchair where the face of the passenger 30 can be seen as shown in FIG. Yes.
Even if the caregiver position detection means 22 loses sight of the caregiver while the wheelchair is temporarily away from the caregiver and heading for the destination, the passenger is highly likely to see the caregiver.
Therefore, the caregiver position detection means 22 that has lost sight of the caregiver compares the face direction of the passenger detected by the passenger face direction detection means 28 with the direction of the front face detected by the front face detection means 27, and boarding When the front face detection means 27 detects the front face in the direction of the person's face, the front face is regarded as a caregiver and tracking is resumed.
In addition, since there is a high possibility that the destination is the destination in the direction in which the passenger faces his face, the movement path determination unit 23 determines whether the passenger's face direction detected by the passenger face direction detection unit 28 and the surroundings From the surrounding situation detected by the situation detection means 21, an object to be noticed by the occupant is identified, and the movement route is selected using it as a landmark of the destination.

In addition, the movement path determination unit 23 provides care for the passenger's face direction detected by the passenger face direction detection unit 28 and the caregiver's face direction detected by the caregiver face direction detection unit 26. It is determined that the carer and the passenger are talking, information on the body direction of the carer 40 is obtained from the carer body detection means 24, and a route for running parallel to the carer is selected so that the user can easily talk. To do.
Also in this case, as shown in FIG. 17, a pan / tilt zoom camera 51 is separately provided at a position where the passenger's face is reflected, and the pan / tilt zoom camera 51 is positioned at the position of the passenger's face detected by the passenger's face orientation detection means 28. The passenger's face direction detection means 28 detects the orientation of the passenger's face with high accuracy from the face image of the passenger reflected in the pan / tilt zoom camera 51. May be.

(Sixth embodiment)
As shown in FIG. 18, the autonomous mobile wheelchair according to the sixth embodiment includes a moving body tracking unit 29 that tracks a moving object in addition to the configuration of the fifth embodiment (FIG. 15).
The hardware configuration of this autonomous mobile wheelchair is the same as that of the first embodiment (FIG. 3), and the mobile body tracking means 29 analyzes the information of the sensor mechanism 14 to detect and track the mobile body. It is realized by.
The moving body tracking unit 29 accumulates the movement history of the detected moving body in the memory 11 and predicts the future movement of the moving body using prediction means such as a Kalman filter based on the history.
The movement path determination unit 23 selects the movement path of the wheelchair so as not to collide with the moving body in consideration of the movement of the moving body predicted by the moving body tracking unit 29. Thus, by predicting the movement of the moving body, a planned and smooth avoidance operation can be performed.
In addition, as shown in FIG. 19, the laser side area sensor 42 for tracking a moving body is separately installed at a position where the front of the wheelchair can be easily observed, and tracking of the moving body identified from the information of the sensor mechanism 14 is performed. The sensor 42 may be used for specialization. FIG. 20 shows a hardware configuration of the autonomous mobile wheelchair in this case.

(Seventh embodiment)
As shown in FIG. 21, the autonomous mobile wheelchair of the seventh embodiment includes a movement direction presentation means 31 that presents the movement direction of the wheelchair in addition to the configuration of the sixth embodiment (FIG. 18).
As shown in FIG. 22, this autonomous mobile wheelchair includes an information presentation device 43 such as a laser pointer or a projector as a hardware configuration, and the movement direction presentation unit 31 determines the traveling direction selected by the movement route determination unit 23. This is realized by the processing of the CPU 17 displayed on the information presentation device 43.

For example, as shown in FIG. 23, the moving direction presentation means 31 displays the traveling direction of the wheelchair on the road surface with an arrow or the like using the laser pointer of the information presentation device 43 attached to the tip of the wheelchair.
Therefore, the surrounding people can act so as not to disturb the progress of the wheelchair, and the wheelchair can move smoothly toward the destination.

  The present invention can be widely used for wheelchairs used in various places such as hospitals, homes, exhibition facilities such as museums, movie theaters, and theaters.

DESCRIPTION OF SYMBOLS 10 Autonomous mobile wheelchair 11 Memory 12 Motor control board 13 Wheel motor 14 Sensor mechanism 15 Laser side area sensor 16 Omnidirectional camera 17 CPU
DESCRIPTION OF SYMBOLS 21 Ambient condition detection means 22 Caregiver position detection means 23 Movement path discrimination means 24 Caregiver trunk detection means 25 Movement control means 26 Caregiver face direction detection means 27 Front face detection means 28 Passenger face direction detection means 29 Mobile body tracking means DESCRIPTION OF SYMBOLS 30 Passenger 31 Movement direction presentation means 40 Caregiver 42 Laser side area sensor for tracking a moving body 43 Information presentation apparatus 50 Pan tilt zoom camera 51 Pan tilt zoom camera

Claims (9)

A wheelchair that moves autonomously according to the caregiver's movement,
Sensor means for providing information representing the surrounding situation;
Caregiver position detecting means for detecting the position of the caregiver based on information of the sensor means;
Ambient condition detection means for detecting surrounding people and obstacles based on the information of the sensor means,
A caregiver torso detection means for detecting a contour shape obtained by cutting the caregiver's torso based on information on the distance and direction from the sensor means to the caregiver;
A movement path determination means for determining a movement path of the wheelchair based on the detection results of the caregiver position detection means , the surrounding situation detection means and the caregiver torso detection means ;
Equipped with a,
The caregiver torso detection means identifies the orientation of the caregiver's body from the detected substantially elliptical contour shape,
The autonomous mobile wheelchair characterized in that the movement route determination means determines the movement route in consideration of information on the direction of the caregiver detected by the caregiver torso detection means . The autonomous mobile wheelchair according to claim 1,
The sensor means comprises a combination of an omnidirectional camera and a distance sensor, and provides image information in which information on the distance and direction to the subject is associated with a subject captured by the omnidirectional camera. An autonomous mobile wheelchair.   3. The autonomous mobile wheelchair according to claim 1, wherein the movement route determination unit determines the movement route while interweaving a route that follows the caregiver and a route that guides the caregiver. Features an autonomous mobile wheelchair. A autonomous mobile wheelchair according to claims 1 to 3, further comprising a caregiver orientation detecting means for detecting the orientation of the face of the caregiver information on the basis of said sensor means, said movement path determining means The autonomous mobile wheelchair is characterized in that the moving route is determined in consideration of the detection result of the caregiver face direction detecting means. A autonomous mobile wheelchair according to any one of claims 1 to 4, further comprising a passenger orientation detecting means for detecting the orientation of the face of the occupant on the basis of the information of said sensor means, the moving path An autonomous mobile wheelchair characterized in that the determining means determines the moving route in consideration of the detection result of the passenger face direction detecting means. The autonomous mobile wheelchair according to any one of claims 1 to 5 , further comprising a front face detecting means for detecting a front face facing the direction of the wheelchair based on information of the sensor means, and the movement path An autonomous mobile wheelchair characterized in that the determining means determines the moving route in consideration of the detection result of the front face detecting means. The autonomous mobile wheelchair according to any one of claims 4 to 6 , further comprising a zoom camera, wherein the caregiver face direction detecting means, the passenger face direction detecting means or the front face detecting means is the zoom camera. An autonomous mobile wheelchair characterized by confirming the face direction of the caregiver, the face direction of the occupant, or the front face using the image. The autonomous mobile wheelchair according to any one of claims 1 to 7 , further comprising a moving body tracking unit that tracks the movement of the moving body, wherein the moving path determination unit displays the tracking result of the moving body tracking unit. An autonomous mobile wheelchair characterized by determining the travel route in consideration. 9. The autonomous mobile wheelchair according to any one of claims 1 to 8 , further comprising information presenting means for presenting the traveling direction of the wheelchair determined by the travel route determining means.