KR20190024467A - Method of generating local path by target orientation and robot implementing thereof - Google Patents

Abstract

The present invention relates to a target orientation local path generation method and a robot for implementing the same. A robot dividing-working a space according to an embodiment of the present invention includes a map storage unit storing a global path including multiple target cells from a start point to a target point and the position of a fixed object in the space of robot movement and a local path generated by reflection of the result of obstacle sensing by a sensing unit in correspondence to the direction and distance of robot movement to at least one of N target cells selected based on the current robot position in the target cell disposed on the global path, and a control unit controlling the sensing unit, the map storage unit, a moving unit, and a working unit, selecting on a priority basis the target cell corresponding to the direction and distance of non-obstacle sensing, and generating the local path of robot movement to the selected target cell.

Description

METHOD OF GENERATING LOCAL PATH BY TARGET ORIENTATION AND ROBOT IMPLEMENTING THEREOF FIELD OF THE INVENTION [0001]

The present invention relates to a method for generating a target-oriented local path and to a robot implementing the method.

In a large-scale floating population space where human and physical exchanges such as airports, schools, government offices, hotels, offices, factories, gymnasiums, and theaters are actively occurring, robots can move based on information about a given space. In this process, the robot maintains the information of the target point to be reached and can create and move the path. That is, the robot can generate a path from the current position to the target point (target point), and can move based on the generated path.

However, various variables can be generated in the process of moving based on the time and route of generating the route. The variables that can occur are the obstacles that are not predicted by the robot on the path generated by the robot. Or the structure of the space at the time when the map information is generated and when the robot travels may be changed.

In order to correspond to the variable, the robot can generate a local path reflecting the local information when generating the path. Particularly, in the large-scale floating population space where the above-mentioned human and material exchanges are active, a situation occurs in which a person moves or obstacles are placed even after the robot creates a path. Therefore, we will consider how to effectively generate a local path and reflect it in the course of travel in order to cope with the same variables as the situation where obstacles are placed in the process of creating and moving the robot.

In order to solve the above problems, the present invention proposes a method of estimating and generating a global path and a local path in a large area, and a robot implementing the method.

Also. In this specification, a robot and a method for traveling along a real-time optimal path while avoiding an obstacle by moving a robot by integrating a global path and a local path are provided.

The objects of the present invention are not limited to the above-mentioned objects, and other objects and advantages of the present invention which are not mentioned can be understood by the following description and more clearly understood by the embodiments of the present invention. It will also be readily apparent that the objects and advantages of the invention may be realized and attained by means of the instrumentalities and combinations particularly pointed out in the appended claims.

A method of generating a target-oriented local path according to an exemplary embodiment of the present invention includes a step of generating a global path including a plurality of target cells from a start point to a target point using a fixed map of a map storage unit, And generating the sensing information corresponding to a direction and a distance in which the robot moves in at least one of the N target cells selected by the control unit based on the position of the current robot among the target cells arranged in the global path, Sensing the obstacle not stored in the supplementary fixed map, the control unit selecting the target cell corresponding to the direction and the distance in which the obstacle is not sensed in accordance with the priority, and the control unit selects the local path on which the robot moves with the selected target cell And the control unit includes controlling the moving unit of the robot so that the robot moves along the local path.

A robot for generating a target-oriented local path according to another embodiment of the present invention includes a sensing unit, a moving unit, a work unit, a robot including a plurality of target cells from a starting point to a target point, The global path, and the N target cells selected based on the position of the current robot among the target cells arranged in the global path, and reflects the result of sensing the obstacle by the sensing unit corresponding to the direction and distance that the robot moves A map storage unit, a map storage unit, a moving unit, and a work unit, and the control unit selects a target cell corresponding to a direction and a distance in which no obstacle is sensed according to a priority order, And generates a local path through which the robot moves.

When the embodiments of the present invention are applied, the robot moves by integrating the global route and the local route, so that the robot can travel according to the real time optimal route while avoiding obstacles.

In addition, when the embodiments of the present invention are applied, after the global path is generated, the robot is moved based on the local path in the process of actual movement, thereby improving the obstacle avoidance ability and maintaining the traveling speed of the robot, So that the robot can travel.

The effects of the present invention are not limited to the effects described above, and those skilled in the art of the present invention can easily derive the various effects of the present invention in the constitution of the present invention.

1 is a block diagram of a robot for generating a target-oriented local path according to an embodiment of the present invention.
FIG. 2 and FIG. 3 are diagrams showing a configuration of a fixed map stored in a map storage unit according to an embodiment of the present invention.
3 is a diagram showing a configuration of a fixed map in which the positions of objects sensed by various sensing parts of the robot are displayed.
4 is a diagram illustrating a process in which a robot generates a global path and generates and moves a local path according to an embodiment of the present invention.
5 is a diagram illustrating a process of generating a global path according to an embodiment of the present invention.
FIG. 6 is a diagram illustrating a process of generating a local path according to an exemplary embodiment of the present invention. Referring to FIG.
FIG. 7 is a diagram for sensing the presence or absence of an obstacle within a certain range and comparing the sensing region with a target cell according to an embodiment of the present invention.
FIG. 8 is a diagram for sensing an obstacle within a certain range of a sensing unit according to another embodiment of the present invention and comparing the same with a target cell.
FIG. 9 is a diagram for determining a traveling direction of a robot based on sensing results of sensing units according to an embodiment of the present invention.
FIG. 10 is a diagram illustrating a process of moving a robot without generating a local path, unlike FIG.
FIG. 11 is a view showing a result of sensing the sensing part of the robot by level to generate the local path of the present invention.
12 is a diagram illustrating movement of a robot combining a global path and a local path according to an embodiment of the present invention.
FIG. 13 is a diagram illustrating a robot selecting a specific sensing area and reflecting it on a local path according to an embodiment of the present invention.
FIG. 14 is a view showing a robot selecting a specific sensing area and reflecting it on a local path according to another embodiment of the present invention.
15 is a diagram illustrating a process of generating a target-oriented local path by a robot according to an embodiment of the present invention.
16 is a diagram illustrating a process of generating a global path and a local path according to another embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings so that those skilled in the art can easily carry out the present invention. The present invention may be embodied in many different forms and is not limited to the embodiments described herein.

In order to clearly illustrate the present invention, parts not related to the description are omitted, and the same or similar components are denoted by the same reference numerals throughout the specification. Further, some embodiments of the present invention will be described in detail with reference to exemplary drawings. In the drawings, like reference numerals are used to denote like elements throughout the drawings, even if they are shown on different drawings. In the following description of the present invention, a detailed description of known functions and configurations incorporated herein will be omitted when it may make the subject matter of the present invention rather unclear.

In describing the components of the present invention, the terms first, second, A, B, (a), (b), and the like can be used. These terms are intended to distinguish the components from other components, and the terms do not limit the nature, order, order, or number of the components. When a component is described as being "connected", "coupled", or "connected" to another component, the component may be directly connected or connected to the other component, Quot; intervening "or that each component may be" connected, "" coupled, "or " connected" through other components.

The present invention may be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art. As shown in FIG.

Hereinafter, terms to be used in this specification will be defined. The global path means a path from the starting position or the current position of the robot to the target point by using a starting point.

The local path refers to a path newly created by the robot in the course of traveling along a global path or a newly created local path and is a path that the robot detects and generates surrounding obstacles with a certain period .

1 is a block diagram of a robot for generating a target-oriented local path according to an embodiment of the present invention. A sensing unit 100 for sensing an object (i.e., an obstacle) disposed outside as a component of the robot 500, a map storage unit 200 for storing a position of a fixed object in a space in which the robot moves, A communication unit 300 for exchanging information with an external server or another external robot,

And a work unit 320 for performing a work by the robot. And a control unit 400 for controlling the respective components, and an interface unit 330 for outputting the operation state of the robot and receiving an instruction input from the outside. According to the embodiment, the communication unit 300 and the interface unit 330 may be optionally included in the robot or not.

The sensing unit 100 refers to various types of sensors that sense the presence, distance, and characteristics of an external object. When the embodiments of the present invention are applied, the sensing unit 100 includes a depth sensing unit for calculating a depth value, a vision sensing unit for calculating vision information, an infrared (IR) sensing unit, an ultrasonic sensing unit, . ≪ / RTI > The robot may more accurately detect an external obstacle by using two or more sensing parts.

In one embodiment of the present invention, the sensing unit 100 mainly includes an external object (such as an object or a robot, such as a person or a person) disposed outside using the RIDOS sensing unit 110 and the depth sensing unit 120, An obstacle that may collide). For example, it is possible to combine the various sensing units described above. For example, the sensing unit 110 and the vision sensing unit, the infrared sensing unit and the depth sensing unit 120, the ultrasonic sensing unit and the lidar sensing unit 110, , Or each of the sensing units can be used independently to sense an external object.

The Lidar sensing unit 110 may sense an object of a specific height and the depth sensing unit 120 may sense objects of various heights. However, since the depth sensing unit 120 has a shorter recognition range than the Rada sensing unit 110, it shows an embodiment of fusing two sensor data.

According to another embodiment of the present invention, it is possible to combine the result of sensing by the ladder sensing unit 110 and the result of sensing by the depth sensing unit 120 to generate a two-dimensional or three-dimensional map. In this process, an infrared sensing unit may be used instead of the lidar sensing unit 110. For example, a plurality of infrared sensing units may be disposed, infrared rays may be transmitted at a certain angle, and an object may be sensed in a point manner, such as the Rada sensing unit 110, according to characteristics of received signals.

In addition, the infrared sensing unit can be used instead of the depth sensing unit 120. The infrared sensing unit may be disposed at various heights of the robot 500 so that height information of objects in a forward direction or objects in a certain distance may be sensed among the objects sensed by the sensing unit 110, The information of the sensed objects and the information sensed by the sensing unit 110 may be combined to generate a two-dimensional or three-dimensional map.

The map storage unit 220 of FIG. 1 may store the global path and the local path in the global path storage unit 210 and the local path storage unit 220, respectively.

FIG. 2 and FIG. 3 are diagrams showing a configuration of a fixed map stored in a map storage unit according to an embodiment of the present invention. The position of the obstacle in the entire space can be identified as 11 in Fig. Then, dividing the entire space by a predetermined cell results in 12 in Fig. If the values of the cell with the obstacle and the cell without the obstacle are set differently based on 12 of FIG. 2, it can be a fixed map.

Of course, in addition to the cell value, you can also display obstacles in black, such as 12. The color display includes displaying the value of the cell stored in the interface unit of the robot as a specific cell. When the robot is stored in the map storage unit of the robot, a value indicating an obstacle placed in each cell can be stored have.

On the other hand, when two or more sensing units are used to store the results of sensing the obstacles on the map, they can be stored separately. For example, an object sensed by the Rada sensing unit 110, an object sensed by the depth sensing unit 120, and an object sensed by both the Rada sensing unit 110 and the depth sensing unit 120 are classified into a map And can be stored in the storage unit. This will be described with reference to FIG.

3 is a diagram showing a configuration of a fixed map in which the positions of objects sensed by various sensing parts of the robot are displayed. The fixed map 200a is stored in the map storage unit 200, and the robot 500 may have at least one fixed map 200a. FIG. 3 is a map showing whether a fixed object that does not move based on a cell to which the entire space is uniformly distributed is arranged.

The size of the cell in FIG. 3 can be determined in consideration of the moving speed of the robot, the size of the robot, or the sensing resolution of the sensing unit 100. For example, the size of the cell can be determined based on the size of the robot, and the size of the cell can be determined by the moving speed per second based on the moving speed of the robot.

Objects (obstacles) in the fixed map 220a of FIG. 3 may be represented by D / L / F. In the case of "D", this means that the depth sensing unit 120 senses an object in the corresponding cell. In the case of "L", the sensing unit 110 senses an object, Means that the object is sensed in the corresponding cell. In case of "F ", both the lidar sensing unit 110 and the depth sensing unit 120 sense the object.

The map configuration of FIG. 3 corresponds to an embodiment. The fixed object may be displayed as a categorized number according to the configuration method, or the color of the cell may be specified so that the object can be displayed and the position of the object can be confirmed.

3, the robot can generate a global path to move to a target point (Goal Point, G) with a specific position as a starting point. In particular, the robot can create a global path according to the type of work the robot should perform. In FIG. 3, the global path generated by the controller 400 of the robot 500 is shown by a line.

For example, it shows the path from the (0, 0) to the global path reaching (18, 13). In this process, the global path can be generated in consideration of the type of work to be performed by the robot 500 or the position (P) of the cell that the robot must necessarily include in the path. Therefore, FIG. 3 shows the global path including the pass point (P), which is a point that must be included in the global path up to the target point G of the robot, in the form of arrows.

3, when the global path is generated based on the fixed map 200a, the positions of the obstacles whose existence is not confirmed in the moving objects, that is, the fixed map 200a, are reflected in the generation of the global path It does not. This is because the controller 400 of the robot can consider only the obstacles (D / L / F) on the fixed map 200a in generating the global path. Therefore, when sensing the obstacle that was not present in the fixed map 200a in the movement process, the controller 400 of the robot generates a new local path and can move based on the new local path.

On the other hand, the local path can be created in various ways. In the present specification, it is desired to configure the local path to be efficiently generated in accordance with the global path as much as possible.

4 is a diagram illustrating a process in which a robot generates a global path and generates and moves a local path according to an embodiment of the present invention. 4 is a diagram illustrating a process of generating a local path in a process of moving a robot after creating a global route before moving.

The control unit 400 generates a global path by combining the positions of the objects in the space stored in the map storage unit 200 and the points at which the robot 500 should pass through the traveling process (S21). The generation of the global path will be described later. The global path can vary depending on the size of the space, the number of objects, and the number of points that must pass. The creation of the global path can be performed only once before the robot arrives at the destination.

Alternatively, the creation of the global path may be newly created while passing through the points to be passed as described above. Alternatively, the control unit 400 can determine whether the robot is moving at the time of global path generation in accordance with a preset reference (time, moving distance, etc.).

Then, the robot senses the object in the N cells arranged on the path before the movement (S22). As shown in FIG. 3, the cells indicated by the arrows are cells arranged on the global path. The sensing unit 100 is checked to see if the object is located at each cell. N can be changed according to the moving speed of the robot and the distribution situation of the object on the outside.

In one embodiment, if the rate at which a robot moves through one cell is Vc and the probability of an external object appearing is Obj_ratio, N may be proportional to Vc and inversely proportional to Obj_ratio. That is, when the speed Vc of the robot is fast, the time to stay in one cell is shortened, and more cells can be moved within a given time, so that it is possible to detect whether an obstacle is arranged to a further cell can do. On the other hand, if the probability of an object is high, the probability that a new object will appear on the path soon after sensing an object that is an obstacle to a far-away cell can be reduced.

N may be fixedly set in the control unit 400 in advance. On the other hand, according to one embodiment, the controller 400 increases the number of N by integrating the number of objects, the number of occurrences, or the distribution of objects around the obstacle in the process of repeating the process of step S22 Or reduced.

In summary, the value of N is calculated by using the control unit (e.g., a robot) using any one or more of the information necessary for the robot to reach the target cell, the distribution of obstacles, the speed of occurrence of the obstacle, the size of the robot, 400 can determine.

The control unit generates a local path by reflecting the result of sensing the object in the N cells (S23). For example, if no objects are sensed in all of the N cells, the robot can generate a local path that goes directly to the Nth cell. On the other hand, when the object is sensed in some cells among the N cells, the local path can be generated so that the object moves to the cell located farthest away from the sensor.

Steps S22 and S23 are summarized as follows. The control unit 400 plans the local path by using the cells of N (T1, T2, T3, ..., TN) received from the global path and allocates the local path from the TN (The object is not sensed) from the cell TN having a higher priority, and moves to the corresponding cell if the cell is movable.

Then, when the local path is generated, the robot moves according to the generated path, and moves in the course of performing the obstacle avoiding motion (S24). The sensing unit 100 continuously senses the proximity obstacle in the moving process and performs motion to avoid the proximity obstacle when it is sensed.

That is, the avoidance motion of the near obstacle includes a corresponding motion to the obstacle that suddenly appears in the course of movement according to the local path. In one embodiment of the corresponding motion, it is possible to perform a motion of running a wall running along the side surface of the obstacle, or a motion of avoiding the obstacle by rotation after the backward movement.

The proximity obstacle avoidance motion is divided into two methods, namely, the month following method and the back method. In the wall following system, when an obstacle is sensed, the controller 400 controls the robot to move around the obstacle. This includes a method in which, when the width of the sensed obstacle is small, the robot travels along the local route to depart from the local route, and then moves along the local route after being distanced from the obstacle.

On the other hand, the back moving method is a method in which the controller 400 moves backward to move away from an obstacle and then moves in a direction in which there is no obstacle. This is because if the width of the sensed obstacle is large, the speed may be slowed down when the robot follows the walking obstacle. Therefore, the controller 400 controls the back moving method Can be selected.

According to the embodiment of the present invention, the above-described S22 and S23 can be repeated in real time. This can be repeated in real time or at a constant period (for example, 20 ms) in consideration of the moving speed of the robot and a situation in which the object, which is an obstacle, is sensed by the sensing unit 100.

5 is a diagram illustrating a process of generating a global path according to an embodiment of the present invention. The creation of the global path can be performed by the control unit 400 as an environment variable, such as the gateway point P on the map, the target point, and the work to be performed by the robot.

A global path can be created in real time or at a certain time interval (5 to 8 seconds). In addition, the global path can be generated by reflecting the moving speed of the robot, the probability that the objects not arranged in the map are sensed, or the distribution of the sensed objects.

The control unit of the robot confirms the position of the object arranged from the current position to the target point G or the gateway point P to be the intermediate target in the fixed map (S31), and generates the global path by reflecting the environment information of the robot (S32). Here, the environment information refers to a type of work the robot has to perform, a condition that the robot must cover a certain range of cells in the space of the given map, or the like.

For example, if the robot determines that the robot has performed tasks (monitoring, security search, cleaning, etc.) up to one cell on each side of the cell where the robot moves, the robot reflects this in the process of generating the global path, You can create a global path with more than one space.

That is, when the working radius of the robot is larger than the cell, or when the area to be covered by the robot in the space is predetermined (the condition of moving 30% of all cells, etc.) And the size at which the speed is reduced are all environment information, and the controller 400 of the robot can generate the global path by reflecting the environment information.

If the speed at which the robot is reduced in the direction change is large, the controller 400 of the robot can generate a global path so as to reduce the maximum change in direction. This also applies when creating a local path. The direction change is an embodiment, and if the velocity is important during the movement of the robot, the global path can be generated in the direction of maximizing the speed. Also, if it is necessary to cover a lot of space, the control unit 400 can densely generate a global path.

In summary, the environmental information that the robot reflects in the process of creating the global path or the local path may be the moving speed of the robot, the turning speed of the robot, the working cell of the robot, . 5, the controller 400 may generate a global path that passes through the gateway point P and reaches the target point G as shown in FIG. Also, the global path means a path including a plurality of target cells from a position of a fixed object in a space in which the robot moves and a target point to a starting point.

FIG. 6 is a diagram illustrating a process of generating a local path according to an exemplary embodiment of the present invention. Referring to FIG. 41 of FIG. 6 shows an embodiment of the global path. At the starting point (S), the target point (including the gate point) is indicated by G. The order of movement of the global path from S to G is indicated by T1, T2, ..., T10. In FIG. 6, it is assumed that N is set to 3 in N cells received by the robot to generate a local path. The value of N can be changed in the process of moving the global path according to the characteristics of the external obstacle or the moving speed of the robot, which are identified in the process of moving the robot. For example, in an area where a large number of obstacles are sensed, the robot sets the N value to a low value. On the other hand, in the area where the obstacle is hardly sensed, the robot sets the N value to a high value. For this purpose, the number of obstacles (area) placed in a specific sensing area at the time of generating the local path is temporarily stored, and when the next local path is generated, the number of obstacles sensed (area) The value of N can be increased or decreased.

The control unit 400 of the robot sets cells of N = 3, that is, T3, T2, and T1, as target cells at the position of S. The sensing unit 100 senses whether an obstacle object is disposed between T3 and S, which are the farthest target cells. If there is no obstacle between the sensing results S-T3, S creates a local path to go directly to T3. The robot moves according to the generated local path in a linear direction from S to T3 as shown at 42 and reaches T3 without passing through T1 and T2, and T3 becomes a new starting point.

If there is an obstacle between the S-T3s, the sensing unit 100 senses whether an obstacle object is disposed between the distant target cells T2 and S next. If there is no obstacle between the sensing results S-T2, S creates a local path to go directly to T2. The robot that moves according to the generated local path moves in a linear direction from S to T2 as shown in 43 and reaches T2. If it is not possible to move to T2, the robot moves to T1. In this process, if the object is placed in the T1 direction, it can move to the obstacle avoidance motion.

FIG. 7 is a diagram for sensing the presence or absence of an obstacle within a certain range and comparing the sensing region with a target cell according to an embodiment of the present invention.

The sensing unit 100 senses whether an obstacle is arranged in the direction of the target cell to which the robot is to be moved using the ladder sensing unit 110 or the depth sensing unit 120 described above. In one embodiment, the Lidia sensing unit 110 may sense objects disposed within a certain range (angular range of 360 degrees or less). A process of sensing in the space corresponding to the target cells T1 to T3 of the space arranged as shown in FIG.

At the start point S, the sensing unit 100 of the robot senses the object within a certain range. The area to be sensed can be divided into a distance and a direction by a sensing method. In one embodiment, in the case of the lidar sensing unit 110, objects can be sensed within a certain distance within a range of 270 degrees or 360 degrees. The sensing angle can be adjusted in consideration of the movement of the robot and the characteristics of objects in the space.

The lidar sensing unit 110 senses the object according to any one of the angular references divided into various degrees such as 1 degree, 7.5 degrees, 10 degrees, and 15 degrees by dividing the whole angle, As shown in FIG. In order to increase the accuracy of the sensing, in FIG. 7, the Ridas sensing unit 110 at the S position can sense that some angles overlap. For example, in the portion indicated by 45 in FIG. 7, it can be seen that the Rada sensing unit 110 performs sensing 10 times at the S position.

The LA is the first sensing (1 ^st Sensing) using the sensing unit 110, the area is a state object is not sensed. Similarly, the object is not sensed in the second to fourth sensing areas (2 ^nd to 4 ^th ). These second to fourth sensing areas are identified as target cells T1 (priority level 3) and T2 (priority level 2). In this case, the control unit 400 confirms that T1 and T2 are movable cells.

Similarly la is when the sensing unit 110, the non-sensing an object in the step of sensing a 5 to 10 th sensing area using, in particular, the sixth sensing region (6 ^th Sensing), the priority level is one of the target cell T3 is . Accordingly, when the object is not sensed in the sixth sensing area, the controller 400 of the robot can control the moving unit 310 to linearly move to the T3 cell having the highest priority.

If the object is sensed in the sixth sensing area corresponding to the T3 cell area and the distance between the object and the robot is calculated and the object is disposed between the T3 cell and the current position S of the robot, The control unit 400 can control the moving unit 310 to move the robot to the T2 cell.

The sensing ranges of FIG. 7 may be partially overlapping. 7, the first sensing region and the second sensing region are partially overlapped, and the second, third, ..., ninth and tenth sensing regions are overlapped. This is a function that can be selectively implemented by the robot to increase the accuracy of object sensing.

In summary, the Rada sensing unit 110 can sense the distances of the obstacles arranged in the direction of the N target cells to sense the obstacle. For example, the distances of the obstacles arranged in the T3 direction, the T2 direction, and the T1 direction can be sensed. If the obstacle is disposed at a distance closer than T3 in the T3 direction or in the T3 direction, If deployed, the robot does not move to T3.

On the other hand, if the obstacle is not located at a distance shorter than T2 in the T2 direction, or if it is sensed that an obstacle is not disposed near T2, the robot can move to T2.

That is, the distance and direction between the robot and the target cell can be confirmed by using the fixed map stored in the map storage unit 200, and the lidar sensing unit 110 senses the obstacles arranged in the corresponding direction, can do.

Although FIG. 7 has been described mainly with respect to the Raid sensing part 110, the same effect can be obtained by using the depth sensing part 120. FIG. In particular, the Rada sensing unit 110 senses objects within a very wide range of angles while the depth sensing unit 120 senses objects in the direction of robot movement, thereby complementarily sensing an object.

The depth sensing unit 120 can sense a wider range than the Ridas sensing unit 110 when sensing. For example, the depth sensing unit 120 may sense the depth values of the objects in the third to seventh sensing areas indicated by 45 in FIG. As a result, the control unit 400 can confirm the placement of objects using the value sensed by the ladder sensing unit 110 and the value sensed by the depth sensing unit 120.

Alternatively, the depth sensing unit 120 senses the depths of the objects arranged in the direction of the T3 cell among the movable target cells T1, T2, and T3. Then, the depth sensing unit 120 determines whether movement is possible to the T3 based on the sensing. If possible, the robot 500 moves. On the other hand, if it is impossible to move to T3 after checking the positions of the objects arranged in the direction of the T3 cell, the position of the objects in the direction of the T2 cell may be sensed and then moved to T2 or T1.

That is, the depth sensing unit 120 can generate the depth values of the objects disposed in the front direction of the robot. Particularly, when the depth sensing unit 120 generates the depth values of the obstacles arranged in the direction of the target cells, The controller 400 can select a target cell to be moved. In one embodiment, when the depth sensing unit 120 generates a distance of a distance of the obstacle from the direction T3 to a distance that is longer than the distance T3, the local path can be linearly formed from S to T3.

FIG. 8 is a diagram for sensing an obstacle within a certain range of a sensing unit according to another embodiment of the present invention and comparing the same with a target cell.

51, the sensing unit 100 confirms that the obstacle 1 is disposed in the direction of the target cells T1, T2, T3. That is, an obstacle is arranged in the area of the T3 cell, and it is confirmed that the robot collides with the robot in the progress of the robot. As a result, the control unit 400 of the robot moves to T2, which is the next target cell, not T3.

On the other hand, 52 shows a state in which the obstacle 2 is arranged close to the robot in the process of moving to the T2 cell. When the obstacle 2 is disposed close to the robot, the robot moves away from the obstacle 2 while maintaining a certain distance by using various sensing parts of the sensing part 100, for example, an ultrasonic sensing part, an infrared sensing part, .

After the global path is generated, the robot moves based on the local path in the process of actual movement of the robot, thereby improving the obstacle avoidance ability and making the motion of the robot natural. In particular, if there is an obstacle, it is possible to move to an efficient distance based on a local path rather than a global path. If the robot moves only after setting the node-based global path, not only the motion is unnatural, but also the efficiency of avoiding and moving obstacles may be lowered.

Particularly, when the embodiment of the present invention is applied, it is possible to avoid a neighboring obstacle and generate a local path that does not pass all of the points in the process of moving to points to be reached on the global path. In this case, you can maintain the optimality of the global path by creating a local path based on the global path (by adjusting the value of N).

In particular, since the sensing unit 100 senses whether the robot can move to the position of the target cell (TN ~ T1) derived from the global path and moves to the selected target cell instead of depending only on the local path in response to the appearance of the obstacle, You can create a local path on the basis of the path.

In the present specification, the fixed map uses heterogeneous sensing units of various heights (depth and radius), so that all obstacles at various heights can be identified.

FIG. 9 is a diagram for determining a traveling direction of a robot based on sensing results of sensing units according to an embodiment of the present invention. The sensing unit 100 senses whether an obstacle (object) is disposed within a range of level 1 (Lv1), level 2 (Lv2), or level 3 (Lv3). The level corresponds to the target cells on the global path where the robot moves to the target position.

For example, level 1 is a distance range corresponding to a position of a target cell having a first priority, level 2 is a distance range corresponding to a position of a target cell having a second priority, and level 3 is a distance range corresponding to a target having a third priority Is the distance range corresponding to the position of the cell.

Two obstacles are placed in the target cell range of the first priority. The original movement path of the robot needs to be moved linearly to the target position, but when the local path is applied, it can move between the two obstacles. In this case, since the robot can move without performing the close avoidance motion for the two obstacles, the moving speed of the robot can be increased. After passing through the obstacle, the robot again detects an obstacle, generates a local route, and moves accordingly.

In the method of FIG. 9, the sensing unit 100 determines whether there is an obstacle in the process of moving the robot to the target cell (the target cell having the highest priority) that is farthest within the range to which the robot is to move. At this time, the obstacle can be identified based on the size of the robot, the size of the robot, the size of the obstacle, and the margin in the process of sensing the obstacle by the sensing unit 100.

FIG. 10 is a diagram illustrating a process of moving a robot without generating a local path, unlike FIG. The robot moves according to the global path. When the obstacle is sensed in the movement process, it operates according to the proximity avoidance motion avoiding the obstacle. This lowers the speed of the robot. Also, unlike FIG. 9, since the obstacle is moved closer to the obstacle and the obstacle is avoided again, the moving distance of the robot increases and the moving efficiency of the robot is lowered.

FIG. 11 is a view showing a result of sensing the sensing part of the robot by level to generate the local path of the present invention. The level depends on the number of target cells referenced in the global path at the current position of the robot. When the number of target cells increases, the robot can deal with obstacles more variously, but the amount of computation increases. Also, even if a local path that moves to a target cell at a long distance is generated due to an increase in the number of target cells, a new obstacle may appear in the movement process and a situation may arise in which a close avoidance motion must be performed.

Accordingly, the controller 400 of the robot determines the number of target cells in consideration of the number of objects, i.e., the cumulative number of times the obstacles appear, the distribution of obstacles (which can be calculated using the Raid sensing unit 110) Can be adjusted. The control unit 400 confirms the sensing result of the sensing unit 100 that the path that can be traveled at the level 1 is calculated with the target cell being the farthest as a priority. As shown in FIG. 11, the controller 400 can identify a region where a collision occurs in a level and a non-collision region using the Lada sensing portion 110 and the depth sensing portion 120.

That is, if a route that can be traveled at level 1 is found, the route is selected as a route close to the target position. If there is no route that can be traveled at level 1, searches are performed in order of level 2 and level 3. As a result, as shown in FIG. 9, it is possible to maintain the flexible running avoiding the obstacle from the distance, thereby actively reducing the scan range in the narrow passage.

12 is a diagram illustrating movement of a robot combining a global path and a local path according to an embodiment of the present invention. Reference numeral 61 denotes a global path generated by the control unit 400. If the robot does not create a local path, it moves to S, T1, T2, ..., T5, G.

62 and 63 are embodiments in which a local path is generated using three target cells. 62 is an example in which the sensing result of the sensing unit 100 is confirmed that the robot can travel to the highest level T3 among the three target cells, and a local path in which the robot moves is generated at T3.

63 is the second one among the three target cells, which is able to travel up to a high level T2, and the sensing result of the sensing unit 100 is confirmed, thereby generating a local path in which the robot moves by T2. 62, and 63, various sensing units (ultrasonic waves, infrared rays, etc.) can be used together to cope with the sudden approaching obstacle in the course of the movement of the robot.

In the case of applying the above-described embodiments, the robots move by integrating the global path and the local path, thereby ensuring the optimum path and obstacle avoidance and smooth motion. In the embodiment of FIG. 9, which identifies an obstacle at a long distance to generate a local path, and in the embodiment of FIG. 10 which takes a motion to avoid an obstacle while moving based on a global path to a destination position, Efficiency. The control unit 400 may reflect both the radius and the margin of the robot as environment variables in the process of generating the local path to ensure the reliability of the obstacle avoidance.

In addition, various kinds of sensing units (lidar / depth / ultrasonic waves) constituting the sensing unit 100 are arranged at various heights of the robots so that all obstacles having various heights can be recognized, Enabling the robot to travel.

In particular, the sensing speed and accuracy can be enhanced when the sensing unit 100 senses a focus in a specific range. The controller 400 may sense the object only in the direction in which the target cell is located in the process of sensing the object so that the sensing speed and accuracy of the obstacle can be increased.

In FIG. 12, the three target cells (T1, T2, T3) calculated in the global path constitute a local path depending on the presence or absence of an obstacle. Priorities can be assigned in the order of T2 and T1 based on the farthest target cell, that is, the highest priority target cell T3, and whether the robot can move from the farthest (highest priority) target cell You can verify and create a local path. If all the target cells are unable to move because of obstacles, it is possible to perform the wall-following method or the back-moving method according to the motion corresponding to the obstacle.

In particular, when all target cells are obstructed by an obstacle, the current position of the robot can be moved so that the robot can move in another direction. For example, if the movement to T1, T2, T3 in Fig. 12 is impossible due to an obstacle, the robot moves to a cell indicated by M capable of moving at right angles to the target point G to regenerate the local path The process can be repeated. Alternatively, the robot may move to M and create a new global path.

This difference will be discussed in more detail in FIGS. 13 and 14. FIG.

FIG. 13 is a diagram illustrating a robot selecting a specific sensing area and reflecting it on a local path according to an embodiment of the present invention. In Fig. 13, three target cells are arranged, and the obstacles are arranged far away from the target cell. In this case, the controller 400 of the robot sets the sensing area in the direction of the farthest target cell T3 having the highest priority. The sensing unit 100 of the robot senses the object in the sensing area and moves in the direction T3 according to the local path.

Referring to FIG. 13, position information on three target cells (T1, T2, T3) closest to the current position in the global path in the global path is identified. And registers the position of the obstacles within a range defined by the distance from level 4 (LV1 to LV4) at a certain angle (within 270 degrees or 360 degrees). Then, selectable sensing areas are identified based on the position of the registered obstacle.

The selectable sensing region includes an area in which no obstacle is arranged among the directions in which the target cell is located. In this process, it is checked whether or not there is a sensing area that can be sensed in the vicinity of the position of the farthest target cell T3, and if possible, the sensing unit 100 senses whether an obstacle exists in the sensing area, do.

If an obstacle is placed between T3 and the robot, it is checked whether there is an obstacle-free sensing region in the same manner in the order of T2 and T1. If it is confirmed, the obstacle is sensed by setting the corresponding direction as a fixed sensing region.

If both T1, T2, and T3 are blocked by obstacles, the obstacle is automatically close to the obstacle.

FIG. 14 is a view showing a robot selecting a specific sensing area and reflecting it on a local path according to another embodiment of the present invention. In Fig. 14, three target cells are arranged, and the obstacles are arranged adjacent to the target cell. In this case, the controller 400 of the robot can not move the obstacle in the direction of the farthest, the highest priority target cell T3 and the direction T2, and thus the sensing area is moved in the direction of the next target cell T1 Setting. The sensing unit 100 of the robot senses the object in the sensing area and moves in the direction T1 according to the local path.

6 to 14, a local path refers to a direction and distance in which the robot moves in at least one of N target cells selected based on the position of the current robot among the target cells arranged in the global path, 100 means that the control unit 400 has generated the result of sensing the obstacle.

In addition, the controller 400 selects a target cell corresponding to a direction and a distance in which the obstacle is not sensed according to the priority order, and generates a local path in which the robot moves to the selected target cell. Here, the priority includes setting the priority in the farthest order in a certain range (N).

15 is a diagram illustrating a process of generating a target-oriented local path by a robot according to an embodiment of the present invention.

The control unit 400 of the robot 400 generates the global path including a plurality of target cells from the start point to the target point by reflecting the environment information of the robot using the fixed map of the map storage unit reflecting the position of the fixed object S71. As described above, the environmental information reflects at least one of a target point on a fixed map, a gateway point, a range of cells to be operated when the robot runs, and a moving speed of the robot.

If the robot performs the work while moving the space densely, the global path of the robot can have a dense configuration on the fixed map. On the other hand, if the robot necessarily passes through a specific area of the space, the space becomes a gateway point, and the controller 400 can generate a global path so as to pass through the gateway point optimally.

After generating the global path, the robot performs steps S72 to S74 for generating a local path using the sensing process. In more detail, the sensing unit may be stored in a fixed map corresponding to the direction and distance that the robot moves in at least one of the N target cells selected by the control unit 400 based on the current position of the robot among the target cells arranged in the global path (S72). After selecting three target cells in FIGS. 6 to 14, a process of sensing an obstacle in the sensing unit 100 according to the direction and distance that the robot moves to the target cell having the highest priority has been described.

The controller 400 selects the target cell corresponding to the direction and the distance in which the obstacle is not sensed according to the priority order (S73), and the controller 400 generates a local path on which the robot moves with the selected target cell (S74) . Thereafter, the control unit 400 controls the moving unit 310 of the robot so that the robot moves along the local path. As a result, the robot moves according to the local path (S75).

The sensing step of S72 will be described in detail. The sensing unit 100 can sense the obstacle in the sensing area within the entire sensing angle (270 to 360 degrees), and can sense the obstacle in the specific sensing area. Accordingly, in the sensing step, the controller 400 selects the N closest target cells T1, T2, ..., TN to be moved in the global path based on the current position of the robot. 6 to 14, the process of selecting three target cells T1, T2, T3 by setting N to 3 has been described. The control unit 400 assigns priorities in the order of TN, TN-1, ..., Tl in order of T1, T2, ... in the order close to the position of the robot.

That is, the farthest target cell among the selected target cells is assigned to have the highest priority and the closest target cell is assigned to have the lowest priority. The control unit 400 then compares the target cell with the target cell in order of priority with respect to the position information of the obstacle sensed by the sensing unit 100. If the target cell having the higher priority is able to move, Set to the local path.

As shown in FIG. 6, when the robot sets only the global path, it moves to T1-> T2-> ..., -> T10-> G. However, if the local path is set, the robot can move directly from the S point to T3 without moving to T1 and T2.

When the embodiment of the present invention is applied, a global path and a local path can be estimated and generated in a large area. In addition, since two or more types of sensing units can be used, it is possible to compare the obstacles having various heights to the fixed map or to update the fixed object with the fixed object to generate the global path to the final destination.

Then, the local path is generated using the position information of the target cells on the route (determined by N) within a radius near the current robot position in the created global path. In this case, the obstacles sensed by the two or more sensing units obtained above are abstracted and used for local path planning, as shown in FIGS. 13 and 14. FIG.

Since the robot moves by integrating the global route and the local route, it can travel along the optimum route in real time while avoiding obstacles. Also, since the local path is generated by considering the size of the obstacle, the size of the robot, and the margin that can be passed, the traveling speed can be secured while avoiding obstacles safely.

16 is a diagram illustrating a process of generating a global path and a local path according to another embodiment of the present invention.

The control unit 400 generates a global path by combining information on the positions of the objects in the space stored in the map storage unit 200 and the points at which the robot 500 should pass through the traveling process (S80). The global path can vary depending on the size of the space, the number of objects, and the number of points that must pass. The creation of the global path can be performed only once before the robot arrives at the destination.

Alternatively, the creation of the global path may be newly created while passing through the points to be passed as described above. Alternatively, the control unit 400 can determine whether the robot is moving at the time of global path generation in accordance with a preset reference (time, moving distance, etc.).

Thereafter, the robot senses the object in N target cells disposed on the path before movement (S81). This involves generating more optimized local paths at the time of path creation and also sensing whether new objects that are not reflected in the global path creation process are placed in the path.

Includes moving people, moving objects with people, or sensing newly placed fixed objects (walls, doors, desks, etc.). If a new object is sensed on the sensing result path and the robot needs to move only along the global path, generation of the local path is unnecessary, so the robot moves according to the created global path (S84). On the other hand, if it is determined that the robot needs to generate a local path more efficiently than the global path as a result of sensing in step S82, the robot generates a local path (S83). When the local path is generated, the robot moves according to the generated local path (S85).

If the sensing unit 100 continuously senses the proximity obstacle in the moving process and the proximity obstacle is sensed (S86), the control unit 400 of the robot performs motion to avoid the proximity obstacle (S87). The avoidance motion of the proximity obstacle includes a corresponding motion to the obstacle that suddenly appears in the process of moving the robot along the path after generating the path. In one embodiment of the corresponding motion, it is possible to perform a wall running (wall following method) running along the side surface of the obstacle, or a motion (back moving method) to avoid an obstacle by turning back after the obstacle.

If the proximity obstacle is not sensed (S86), the robot confirms whether it has reached the target point (S88). If it does not arrive, it is determined whether or not it is time to create a global route (S89). If it is not necessary to regenerate the global route, movement is performed in a step S81. On the other hand, if the global path is to be regenerated, the global path is created in step S80 and then moved.

The generation time of the global path can be configured according to the time, distance, or environmental change of the robot. For example, if there is no change in the sensed state after generating the global route in step S80, the robot does not correspond to the time when the global route is generated because the robot maintains the global route and moves.

On the other hand, if the global path needs to be modified on the global path, for example, if the obstacle that can not reach the gateway point is newly sensed, it is necessary to create a global path excluding the gateway point. A new global route can be created when the gateway point (P) to which the robot has to pass is excluded from the travel route due to the appearance of the obstacle.

Also, the robot can select the cell corresponding to the passing point in the process of reaching the target point, and calculate the global path or the local path. For example, in FIG. 3, a gate point indicated by P is a cell that the robot must pass through in the movement process. Likewise, the target point marked G is the robot's final destination for the global path. The robot generates the P point in the process of generating the global path, and can reflect this in the process of generating the local path.

In addition, it is to be understood that the present invention is not necessarily limited to these embodiments, and that all the elements within the scope of the present invention Or may be selectively coupled to one or more of them. In addition, although all of the components may be implemented as one independent hardware, some or all of the components may be selectively combined to perform a part or all of the functions in one or a plurality of hardware. As shown in FIG. The codes and code segments constituting the computer program may be easily deduced by those skilled in the art. Such a computer program can be stored in a computer-readable storage medium, readable and executed by a computer, thereby realizing an embodiment of the present invention. The storage medium of the computer program includes a magnetic recording medium, an optical recording medium, and a storage medium including a semiconductor recording element. Also, a computer program embodying the present invention includes a program module that is transmitted in real time through an external device.

While the present invention has been particularly shown and described with reference to exemplary embodiments, it is to be understood that the invention is not limited to the disclosed exemplary embodiments. It is therefore to be understood that such changes and modifications are intended to be included within the scope of the present invention unless they depart from the scope of the present invention.

100: sensing unit 200: map storage unit
300: communication unit 310:
320: Operation unit 330: Interface unit
400: control unit 500: robot

Claims (13)

Generating a global path including a plurality of target cells from a start point to a target point by reflecting the environment information of the robot using a fixed map of a map storage unit in which the control unit of the robot reflects the position of the fixed object;
Wherein the sensing unit detects an obstacle that is not stored in the fixed map in correspondence with a direction and a distance in which the robot moves in at least one of N target cells selected by the control unit based on a position of the current robot among the target cells arranged in the global path, Sensing;
Wherein the control unit selects the target cell corresponding to the direction and the distance in which the obstacle is not sensed in priority order;
Wherein the controller generates a local path through which the robot moves with the selected target cell; And
Wherein the control unit comprises controlling the moving part of the robot such that the robot moves along the local path.
The method according to claim 1,
The sensing step
Selecting, by the control unit, the nearest N target cells T1, T2, ..., TN to be moved in the global path based on the current position of the robot;
The control unit assigning priorities in the order of TN, TN-1, ..., T1; And
Wherein the control unit includes a step of comparing the target cell with the location information of the obstacle sensed by the sensing unit according to the order of the priority.
The method according to claim 1,
Wherein the sensing unit includes a ladder sensing unit,
The step of sensing the obstacle includes:
Further comprising the step of sensing the distance of an obstacle located in the direction of the N target cells.
The method according to claim 1,
The sensing unit includes a depth sensing unit,
The step of sensing the obstacle includes:
Further comprising generating the depth value of an obstacle disposed in a direction of the N target cells, wherein the depth sensing unit generates a depth value of the obstacle.
The method according to claim 1,
The environmental information
Which is environment information reflecting at least one of a target point on a fixed map, a gateway point, a range of a cell to be operated when the robot runs, and a moving speed of the robot.
The method according to claim 1,
Wherein the controller further comprises setting the N using at least one of a travel time required for the robot to reach the target cell, a distribution of an obstacle, an arrival speed of an obstacle, and a size of the robot, How to create a path.
The method according to claim 1,
The step of controlling the moving unit
Sensing the obstacle proximate to a direction in which the robot moves; And
The control unit may control the moving unit to move the robot according to any one of a wall following method of moving along the periphery of the obstacle or a back moving method of moving back and rotating after the obstacle is sensed To create a target-oriented local path.
A robot for generating a target-oriented local path,
A sensing unit for sensing an obstacle disposed outside;
A global path including a plurality of target cells from a position of a fixed object in a space in which the robot moves and a target point from a start point to a target point, and N target cells selected based on a position of a current robot among target cells arranged in the global path A map storage unit for storing a local path generated by reflecting the result of sensing the obstacle by the sensing unit corresponding to a direction and a distance that the robot moves;
A moving unit for moving the robot;
An operation unit for performing operations of the robot; And
The control unit controls the sensing unit, the map storage unit, the moving unit, and the work unit, and the control unit selects the target cell corresponding to the direction and the distance in which the obstacle is not sensed, And a control unit for generating the local path through which the robot moves.
9. The method of claim 8,
The control unit selects the N closest target cells T1, T2, ..., TN to be moved in the global path on the basis of the current position of the robot and outputs the selected target cells in the order of the target cells TN, TN-1, A target-oriented local path is generated in which the target cell is compared with the position information of the obstacle sensed by the sensing unit in accordance with the order of priority.
9. The method of claim 8,
Wherein the sensing unit includes at least one of a ladder sensing unit and a depth sensing unit,
Wherein the sensing unit generates distance information or depth information of an obstacle placed in a direction of the N target cells, and generates a target-oriented local path.
9. The method of claim 8,
The environmental information
Wherein the target-oriented local path is environment information that reflects at least one of a target point on the fixed map, a gateway point, a range of cells to be operated when the robot runs, and a moving speed of the robot.
9. The method of claim 8,
Wherein the controller further comprises setting the N using at least one of a travel time required for the robot to reach the target cell, a distribution of an obstacle, an arrival speed of an obstacle, and a size of the robot, Robots that generate paths.
9. The method of claim 8,
The control unit controls the moving unit of the robot so that the robot moves along the local path,
When the sensing unit senses an obstacle proximate to the direction in which the robot moves, and the obstacle is sensed by the sensing unit, the control unit may perform a wall-following method of moving along the obstacle or a back- And controlling the moving unit such that the robot moves according to any one of the two directions.

Images