KR20090065890A - Situation dependent behavior based robot control appartus and method - Google Patents

Abstract

Behavior base robot control apparatus and method are provided, which can increase the self-exercise nature of robot by deciding the behavior combination generation and control method of robot. The physical sensor part(10) acquires the video information about environment and senses the proximity degree of the opponent robot or obstacle. The sensor unit(20) outputs the outer situation information about the collision risk. The internal state module(40) recognizes clearly the internal state information of robot. The outer situation evaluation module(50) evaluates the current outer situation by using the outer situation information. The behavior combination generation/adjustment module(60) outputs the control resultant signal. The behavior combination generation/adjustment module(70) outputs the control resultant signal by the control method.

Description

Behavior-based robot control device and its control method {SITUATION DEPENDENT BEHAVIOR BASED ROBOT CONTROL APPARTUS AND METHOD}

The present invention relates to an action-based robot control apparatus and a control method thereof.

Traditional mobile robot technology relies on a model of the environment. In this method, the user gives the robot a map (model) of the remote location where the robot will operate in advance, and the robot obtains information on the surrounding situation using a sensor and compares it with the map that has received the information in advance. It's a way of knowing where you are, building a complex action plan, and then acting on it.

However, this method is like finding a path in a transformed area with the old maps. That is, in reality, it is difficult for the user to know the information about the remote environment in advance, and even if the situation can be known in advance, if there are many objects moving from the remote place, or the area where the environment changes frequently, based on the previously known information. It is not practical to control the robot and to carry out the action after a complicated and time-consuming plan. Thus, the existing model-based robot control has many limitations in solving the realistic problem.

Therefore, in order to overcome the limitations of the existing model-based robot control, behavior-based robot control that originates from observation of living things such as insects is required.

The present invention was devised to solve the above problems, and has a behavior-based robot control having a behavior-dependent behavior-based control architecture whose behavior varies according to the situation of the robot. It is an object of the present invention to provide an apparatus and a control method thereof.

According to an aspect of the behavior-based robot control apparatus according to the present invention for achieving the above object, a physical sensor unit for obtaining the image information about the environment and to detect the proximity of the opponent robot or obstacle, and the physical sensor unit Logical sensor unit for outputting the external situation information about the position and direction of the object in the environment and the risk of collision with the opponent robot or obstacle, an internal state module for recognizing the internal state information of the robot, and the external situation An external situation evaluation module for evaluating a current external situation using information, and an action set and adjustment method selected according to an external situation evaluated by the external situation evaluation module and an internal state of the robot recognized by the internal state module. An action set / adjustment module for outputting a coordination result signal and a collision risk signal from the logical sensor unit Is a situation independent action set / coordination module for outputting an adjustment result signal by the selected situation independent action set and the situation independent coordination method, and the adjustment result signal output from the action set / coordination module or the situation independent action set / coordination module. And a motion controller for generating robot path data, and an actuator for driving the robot according to the robot path data generated by the motion controller.

According to one aspect of the behavior-based robot control method according to the present invention for achieving the above object, (a) the external information on the position and direction of the object in the environment and whether or not the risk of collision with the opponent robot or obstacle, the sensor unit Outputting from; (b) evaluating a current external situation through the external situation information by an external situation evaluation module; (c) recognizing the internal state of the robot by the internal state module; (d) outputting an adjustment result signal by an action set and an adjustment method selected according to the evaluated external situation and the internal state of the robot; (e) robot path data is generated according to the output adjustment result signal, and the robot is driven by the generated robot path data.

According to the present invention, there is provided a behavior-based robot control apparatus having a behavior-dependent behavior-based control architecture in which a behavior varies according to a situation in which the robot is located, and a control method thereof. As there is no need for map information about the surrounding environment, and no complicated and time-consuming action plan is required, it can be applied very efficiently to the environment where the environment changes or changes from moment to dynamic. There is an effect to increase.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. Prior to this, terms or words used in the specification and claims should not be construed as having a conventional or dictionary meaning, and the inventors should properly explain the concept of terms in order to best explain their own invention. Based on the principle that can be defined, it should be interpreted as meaning and concept corresponding to the technical idea of the present invention. Therefore, the embodiments described in the specification and the drawings shown in the drawings are only the most preferred embodiment of the present invention and do not represent all of the technical idea of the present invention, various modifications that can be replaced at the time of the present application It should be understood that there may be equivalents and variations.

1 is a view showing the configuration of the behavior-based robot control apparatus according to the present invention, such a behavior-based robot control device is embedded in the robot to be described below to enable the robot to perform autonomous behavior.

As shown in FIG. 1, the behavior-based robot controller of the present invention includes physical sensors 10, logical sensors 20, short term memory 30, and the like. Internal states module 40, external situation assessor 50, behavior set / coordination module 60, independent behavior set / coordination module 70, motion controller (Motion controller) 80 and actuators (90).

Physical sensors 10 are physical sensors, for example, a color CCD camera for acquiring image information about objects in an environment, and a proximity of a relative robot or an obstacle. Two infrared sensors or the like for sensing may be used.

The logical sensors 20 are conceptual sensors based on the physical sensors 10, and target detection sensors, collision detection sensors, and the like are used. The output information of the logical sensor 20 is external situation information including position and direction information of the detected objects and collision risk detection information between an opponent robot or an obstacle.

The short term memory 30 stores external situation information output from the logical sensors 20.

Internal states module 40 stores the internal states of the robot. Here, the internal state may be set by the user or may be changed according to information extracted from the short term memory 30.

In addition, the change of the state inside the robot by the external environment may implement the personality of the robot as the mood of the person may change according to the change of the weather.

The external situation assessor 50 uses the data stored in the short-term memory 30 to evaluate the current external situation. In addition to the current data, this assessment uses data from one step back. Here, the external situation depends on the surrounding environment including the robot itself.

The behavior set / coordination module 60 is composed of n behavior sets and n coordinations, and evaluates internal states and external conditions of the robot stored in the internal state module 40. Depending on the external situation assessed by (50), either one of the behavior sets and the coordination methods is selected from among the n behavior sets and n coordinations methods. For example, when a behavior set set i is selected, coordination i is automatically selected accordingly, and the result of the adjustment is transmitted to the motion controller 80.

That is, even if the output of the logical sensor is the same, the behaviors of the robot may vary according to the internal state or the external situation of the robot. This setup allows one robot to perform a more intelligent and adaptive task than the other, when there are two robots with the same ability to sense the outside world.

The independent behavior set / coordination module 70 selects independent behavior set and siting independent coordination when a specific behavior is to be performed in an emergency situation such as a robot collision risk. Is passed to a motion controller 80.

That is, when collision risk information of the robot is output from logical sensors 20, independent behavior set and independent adjustment method of the independent behavior set / adjustment module 70 regardless of an external situation. (situation independent coordination) is immediately selected and the adjustment result is transmitted to the motion controller 80. Even in this case, the behavior of the robot may vary depending on the internal state of the robot.

In particular, in the case of such an emergency, the output of the logical sensors 20 operates in the same concept as a hardware interrupt. An example of such an emergency is that a life-saving robot performs a task, and the body of the person to be rescued is significantly affected. If it detects a danger, it may be able to perform emergency rescue actions other than normal rescue actions, or one of a number of reconnaissance robots may detect explosives just before the explosion.

The motion controller 80 generates robot path data according to the adjustment result information transmitted from the action set / adjustment module 60 or the independent action set / adjustment module 70.

The actuators 90 drive the actual robot according to the robot path data generated by the motion controller 80.

FIG. 2 is a diagram illustrating an example of a robot in which the behavior-based robot control apparatus of FIG. 1 is embedded.

As shown in FIG. 2, the robot 100 in which the action-based robot control apparatus of the present invention is embedded has a large upper body and a lower body, and has a structure in which the upper body and the lower body can be operated separately. In particular, the color CCD camera 105 as a physical sensor is mounted on the head of the robot to photograph the surrounding environment.

In addition, a built-in command receiver for receiving a wireless command signal transmitted from the outside inside the robot, and the remaining control-related modules other than the physical sensor and the actuator is external to the robot according to the command signal of the control-related module Can be configured to perform immediately. This will be described in more detail below.

3 is a perspective view of a wirelessly controlled robot flag game machine using an action-based robot according to an embodiment of the present invention, and FIG. 4 is a view showing an example of a behavior set adjustment method according to an embodiment of the present invention. There are two robots and four flags, and allied flags are colored blue and opponents are colored red.

As shown in FIG. 3, the wirelessly controlled robot flag game machine according to an embodiment of the present invention may be a friendly robot 100a, an opponent robot 100b, a friendly flag 200a, 200b, or a relative flag 200c, 200d. And an external control device 400 for wirelessly controlling each of the plurality of obstacles 300 present in the space between the friendly robot 100a and the opponent robot 100b and the friendly robot 100a and the opponent robot 100b. It is configured to include).

First, an example of an action set that is selected when four external situations exist according to a change in the relationship between the robot itself and the surrounding environment will be described.

That is, "1. The robot is facing the opponent's flag and the robot can press the flag, 2. The robot is facing the opponent's flag, the robot can't press the flag, 3. The robot is facing our flag and the robot The situation close to the flag, 4. The robot is facing our flag and the robot is far from the flag. Four sets of external situations and the behavior set selected by the robot's internal state will be described in more detail. Here, the internal state of the robot is set to an attack state or a defensive state by the player, and this internal state may be changed according to information extracted from the short term memory.

Table 1 below shows an action set that is selected regardless of the internal state of the robot in the external situation 1.

 External situation  Internal condition  Logical sensor output  Set of actions     The robot is facing the opponent's flag and the robot can press the flag      Don't care  ① Opp_flag  Touch_the_flag  ② Our_flag  Avoiding_the_flag  ③ Opp_robot  Avoiding_opp_robot  ④ None  Searching

As shown in Table 1, when the external situation is a situation where the robot is facing the opponent flag and the robot can press the flag and is not related to the internal state of the robot (Don't care), the action of pressing the opponent flag (Touch_the_flag), A set of actions including moving (Avoiding_the_flag) away from our flag, avoiding opponent robot (Avoiding_opp_robot), and searching (Searching) is selected.

At this time, the action of pressing the relative flag (Touch_the_flag) corresponds to the case where the output of the logical sensor is ① the relative flag (Opp_flag), and the action of moving away from our flag (Avoiding_the_flag) is the output of the logical sensor ② Our_flag. Corresponds to). In addition, avoiding the opponent robot (Avoiding_opp_robot) corresponds to the case where the output of the logical sensor is (3) opponent robot (Opp_robot), and searching is corresponding to the case where the output of the logical sensor is (4) None. .

In particular, when the output of the logical sensor is ④ None, the robot is near the relative flag, so the robot does not move and searches using only neck degrees of freedom.

On the other hand, Table 2 below shows the behavior set that is selected according to the internal state of the robot in the external situation 2.

 External situation  Internal condition  Logical sensor output  Set of actions           The robot is facing the opponent's flag and the robot cannot press the flag.      Offense  ① Opp_flag  Approaching_to_the_flag  ② Our_flag  Avoiding_the_flag  ③ Opp_robot  Avoiding_opp_robot  ④ None  Wandering      Defense  ⑤ Opp_flag  Waiting  ⑥ Our_flag  Blocking  ⑦ Opp_robot  Blocking  ⑧ None  Searching

As shown in Table 2, when the external situation is a situation where the robot is facing the opponent flag, the robot cannot press the flag, and the internal state of the robot is an offense (Offense) state, approaching to the other flag (Approaching_to_the_flag), A set of actions including moving (Avoiding_the_flag) away from our flag, avoiding opponent robot (Avoiding_opp_robot), and wandering (Wandering) are selected.

At this time, the approach to the other flag (Apporachin_to_the_flag) corresponds to the case where the output of the logical sensor is ① the opponent flag (Opp_flag), and the action to move away from our flag (Avoiding_the_flag) is the output of the logical sensor ② our flag. Corresponds to the case of (Our_flag). In addition, avoiding an opponent robot (Avoiding_opp_robot) corresponds to the case where the output of the logical sensor is ③ opponent robot (Opp_robot), and wandering corresponds to the case where the output of the logical sensor is ④ no detection (None). .

In particular, when the output of the logical sensor is ④ None, the robot moves using the legs and wanders using neck freedom.

On the other hand, if the external situation is that the robot is facing the opponent's flag, the robot cannot press the flag, and the internal state of the robot is Defence, the action is to hold the current position to defend the other robot. , A behavior set including blocking the opponent robot, blocking the opponent robot, and searching is selected.

At this time, the act of sticking to the current position to defend the opponent robot corresponds to the case where the output of the logical sensor is ⑤ the opponent flag (Opp_flag), and the act of defending the opponent robot is determined by the output of the logical sensor. ⑥ Corresponds to our flag (Our_flag). In addition, the blocking of the opponent robot corresponds to the case where the output of the logical sensor is ⑦ the opponent robot (Opp_robot), and the searching action corresponds to the case where the output of the logical sensor is ⑧ no detection (None). do.

In particular, if the output of the logical sensor is ⑧ None, the robot will search using neck freedom.

On the other hand, Table 3 below shows the behavior set that is selected according to the internal state of the robot in the external situation 3.

 External situation  Internal condition  Logical sensor output  Set of actions           Robot is facing our flag and robot is near flag      Offense  ① Opp_flag  Approaching_to_the_flag  ② Our_flag  Avoiding_the_flag  ③ Opp_robot  Avoiding_opp_robot  ④ None  Wandering      Defense  ⑤ Opp_flag  Waiting  ⑥ Our_flag  Blocking  ⑦ Opp_robot  Blocking  ⑧ None  Searching

As shown in Table 3 above, when the external situation is a robot facing our flag, the robot is close to the flag, and the internal state of the robot is an offense, the approach to the other flag (Apporachin_to_the_flag) and our flag A set of actions including moving (Avoiding_the_flag), avoiding an opponent robot (Avoiding_opp_robot), and wandering (Wandering) are selected.

At this time, the approach to the other flag (Apporachin_to_the_flag) corresponds to the case where the output of the logical sensor is ① the opponent flag (Opp_flag), and the action to move away from our flag (Avoiding_the_flag) is the output of the logical sensor ② our flag. Corresponds to the case of (Our_flag). In addition, avoiding an opponent robot (Avoiding_opp_robot) corresponds to the case where the output of the logical sensor is ③ opponent robot (Opp_robot), and wandering corresponds to the case where the output of the logical sensor is ④ no detection (None). .

In particular, when the output of the logical sensor is ④ None, the robot moves using the legs and wanders using neck freedom.

On the other hand, if the robot is facing our flag, the robot is close to the flag, and the robot's internal state is defense, defending the opponent's current position and defending the opponent robot. A set of actions including blocking, blocking of the opponent robot, and searching is selected.

At this time, the act of sticking to the current position to defend the opponent robot corresponds to the case where the output of the logical sensor is ⑤ the opponent flag (Opp_flag), and the act of defending the opponent robot is determined by the output of the logical sensor. ⑥ Corresponds to our flag (Our_flag). In addition, the blocking of the opponent robot corresponds to the case where the output of the logical sensor is ⑦ the opponent robot (Opp_robot), and the searching action corresponds to the case where the output of the logical sensor is ⑧ no detection (None). do.

In particular, if the output of the logical sensor is ⑧ None, the robot will search using neck freedom.

On the other hand, Table 4 below shows the behavior set that is selected according to the internal state of the robot in the case of the external situation 4.

 External situation  Internal condition  Logical sensor output  Set of actions           The robot is facing our flag and the robot is far from the flag.      Offense  ① Opp_flag  Approaching_to_the_flag  ② Our_flag  Avoiding_the_flag  ③ Opp_robot  Avoiding_opp_robot  ④ None  Wandering      Defense  ⑤ Opp_flag  Avoiding_the_flag  ⑥ Our_flag  Approaching_to_the_flag  ⑦ Opp_robot  Approaching_to_opp_robot  ⑧ None  Wandering

As shown in Table 4, when the external situation is the robot facing our flag, the robot is far from the flag, and the internal state of the robot is Offense, the approaching to the other flag (Apporachin_to_the_flag) and our flag A set of actions including moving (Avoiding_the_flag), avoiding an opponent robot (Avoiding_opp_robot), and wandering (Wandering) are selected.

At this time, the approach to the other flag (Apporachin_to_the_flag) corresponds to the case where the output of the logical sensor is ① the opponent flag (Opp_flag), and the action to move away from our flag (Avoiding_the_flag) is the output of the logical sensor ② our flag. Corresponds to the case of (Our_flag). In addition, avoiding an opponent robot (Avoiding_opp_robot) corresponds to the case where the output of the logical sensor is ③ opponent robot (Opp_robot), and wandering corresponds to the case where the output of the logical sensor is ④ no detection (None). .

In particular, when the output of the logical sensor is ④ None, the robot moves using the legs and wanders using neck freedom.

On the other hand, when the robot is facing our flag, the robot is far from the flag, and the internal state of the robot is defensive, moving away from the opponent's flag (Avoiding_the_flag) and moving towards our flag ( Action set including Approaching_to_the_flag, Approaching_to_opp_robot, and Wandering is selected.

At this time, the act of avoiding the counterpart flag (Avoiding_the_flag) corresponds to the case in which the output of the logical sensor is ⑤ the opposing flag (Opp_flag), and the act of approaching toward our flag (Approaching_to_the_flag) is that the output of the logical sensor is ⑥ our flag (Our_flag). Corresponds to the case. Approaching_to_opp_robot also corresponds to the case where the output of the logical sensor is ⑦ Oppp_robot, and wandering corresponds to the case where the output of the logical sensor is ⑧ undetected (None). do.

In particular, if the output of the logical sensor is ⑧ None, the robot will search using neck freedom.

On the other hand, when the output of the logical sensor is output in multiples such as the opponent flag (Opp_flag), our flag (Our_flag), and the opponent robot (Opp_robot), it is necessary to adjust the respective actions accordingly.

For example, when the robot internal state is Offense in Offensive 2 as shown in FIG. 4, the behavior set approaches the flag (Approaching_to_the_flag), away from the flag (Avoiding_the_flag), avoids the opponent robot (Avoiding_opp_robot), It consists of four actions, such as Wandering, where Wandering has the lowest priority.

And, compared to Wandering, the remaining three actions have a higher priority, but the adjustment is performed by the sum of the vectors, not by the priority.

In other words, when both the approaching approach (Approaching_to_the_flag) and the avoiding approaching robot (Avoiding_opp_robot) are selected, the moving point and direction of the robot are determined by adding the vectors converged to the flag and the vectors emitted from the opponent.

As such, the robot basically performs an action according to an action set selected according to an external situation and an internal state of the robot. If necessary, it can be manually adjusted by the external control device 400 of the game machine, or can be automatically adjusted by a separate computer.When the wireless control signal is transmitted to the robot wirelessly, the robot receiving the command moves forward, backward walking, You will immediately perform rotational walking, diagonal walking, and raising your arms.

In particular, in the case of automatic adjustment by computer, the signal processing unit of the computer wirelessly receives an image signal of the surrounding environment photographed from one color CCD camera mounted on the head of the robot and analyzes the transmitted image signal. It will be adjusted automatically.

Accordingly, the robot receiving the wireless command signal of the computer moves the obstacle 300 and the opponent robot 100b while approaching or contacting the upper arm with the opponent flags 200c and 200d, thereby making the color of the opponent flag friendly. The player's opponent controls the robot and raises the flag, and after a predetermined time, the game is decided according to the number of colors of the flag.

Hereinafter, with reference to the accompanying Figures 5 and 6 will be described in more detail with respect to the robot control using a computer or an external control device of the game machine.

5 is a view showing the system configuration of the radio control robot flag game machine according to an embodiment of the present invention, Figure 6 is a view showing the structure of the flag unit of FIG.

As shown, the wireless control robot flag game machine system of the present invention is an external control device for manually wirelessly adjusting the robot unit 100, the flag unit 200, and the robot unit 100 located in the playing field of the game machine ( 400 and a computer 500 for automatically wirelessly adjusting the robot unit 100. Here, the computer 500 may be included in the radio controlled robot flag game machine or may be separately configured.

The robot unit 100 is a robot which is manually adjusted by the external adjusting device 400 or wirelessly controlled by the computer 500, and includes a command receiving unit 101, a motor control unit 102, and a motor driving unit 103. And the power supply unit 104 and the vision sensor unit 105.

The command receiving unit 101 receives a command signal transmitted wirelessly from the external adjusting device 400 or the computer 500.

The motor control unit 102 outputs a motor driving signal for controlling the robot by the command signal transmitted from the command receiving unit 101.

The motor driver 103 drives the motor according to the motor drive signal output from the motor controller 102.

The power supply unit 104 supplies power for the robot to operate.

The vision sensor unit 105 is a color CCD camera mounted to the head of the robot, and wirelessly transmits an image signal of the surrounding environment photographed from the color CCD camera to the computer 500, and the computer 500 transmits the The robot is automatically adjusted by analyzing the image signal.

The flag unit 200 is operated by the proximity or contact of the robot, and includes a sensor unit 201, a motor control unit 202, a motor driving unit 203, a motor 204, and a flag support unit 205. It is configured by.

The sensor unit 201 determines a state of contact or proximity of the robot arm. That is, a small pressure sensor capable of detecting the contact of the robot arm or a pair of infrared sensors capable of determining the proximity of the robot arm is used.

The motor controller 202 outputs a motor driving signal for driving the motor 204 when a contact or proximity detection signal of the robot arm is transmitted from the sensor unit 201.

The motor driver 203 drives the motor 204 according to the motor drive signal output from the motor controller 202.

The flag support 205 is two flag support members 205a and 205b attached to the rotational axis of the motor 204. The flag support members 205a and 205b have different colored flags 200a and 200c at the ends of the flag support members 205a and 205b. Is attached.

That is, when the motor 204 is driven by the motor driving unit 203, the two flags 200a and 200c attached to the rotating shaft of the motor 204 are switched, so that the flag 200a in the vertical direction is connected to the robot. Visible to the player, the other flag 200c enters into the body of the flag portion and becomes invisible.

External control device 400 is a device for the player to manually adjust the robot, and transmits a radio command signal input by the player to the robot, in particular, the player using the external control device 400, the internal state of the robot ( Attack state, defense state).

The computer 500 is a device for automatically adjusting a robot by a player. The computer 500 wirelessly receives an image signal of a surrounding environment captured by a color CCD camera mounted on the head of the robot, and analyzes the transmitted image signal. The robot will be adjusted automatically.

That is, the player wirelessly controls the robot in the playing field through the external control device 400 or the computer 500 to avoid the opponent robot, and uses the robot arm to contact or approach the opponent flag by using his robot arm. The game progresses by switching to or defending against an opponent robot approaching his flag. After a certain time, the game is decided by the number of colors of the flag.

7 and 8 are diagrams illustrating a behavior-based robot control method according to an embodiment of the present invention.

As shown in the figure, the position / direction information of the opponent robot and the flag detected by the physical sensor and the collision risk information between the opponent robot or the obstacle are transmitted to the logical sensor (S10).

Subsequently, the logical sensor checks the information transmitted from the physical sensor to determine whether an emergency situation such as a collision risk has occurred (S20), and when the collision risk does not occur, output information of the logical sensor is stored in the short-term memory (S30). .

Subsequently, the current external situation is evaluated based on the output information of the logical sensor stored in the short-term memory (S40), and one of the n action sets and the n adjustment methods is adjusted according to the evaluated external situation and the internal state of the robot. This is selected (S50).

Subsequently, the robot path data is generated according to the adjustment result by the selected action set and the adjustment method (S60), and the actual robot is driven (S70) according to the generated robot path data.

In particular, in the process of S50 in which any action set and adjustment method are selected from the n action sets and the n adjustment methods according to the evaluated external situation and the internal state of the robot, the external situation is described as described above. A situation where the robot is facing the flag and the robot can press the flag, 2. the robot is facing the opponent's flag and the robot cannot press the flag, 3. the robot is facing our flag and the robot is close to the flag, 4. the robot There may be four external situations facing the opponent's flag and the robot is far from the flag. In these four external situations, the internal state of the robot and the output of the logical sensor for the external situations 1 and 2 Since the behavior set selected according to the foregoing has been described above, a detailed description thereof will be omitted.

Meanwhile, in the process of S20 of determining whether an emergency situation such as a collision risk has occurred by checking the information transmitted from the physical sensor in the logical sensor, if it is determined that there is a collision risk, an independent action set according to the collision risk and an independent adjustment method This is selected (S80).

Subsequently, the robot path data is generated according to the selected independent action set and the adjustment result by the independent adjustment method (S90), and the actual robot is driven (S100) according to the generated robot path data.

In other words, in case of an emergency such as a robot collision risk, the robot's output is generated according to the result of the adjustment by selecting the independent action set and the independent adjustment method immediately because the output of the logical sensor operates in the same concept as the hardware interrupt regardless of the external situation. The actual robot is driven by the path data, and even in this case, the behavior of the robot may vary according to the internal state of the robot.

As described above, although the present invention has been described by way of limited embodiments and drawings, the present invention is not limited thereto and is intended by those skilled in the art to which the present invention pertains. Of course, various modifications and variations are possible within the scope of equivalents of the claims to be described.

1 is a view showing the configuration of the behavior-based robot control apparatus according to the present invention.

FIG. 2 is a diagram illustrating an example of a robot in which an action-based robot controller of FIG. 1 is embedded.

Figure 3 is a perspective view of a radio control robot flag game machine using an action-based robot according to an embodiment of the present invention.

4 is a diagram illustrating an example of a behavior set adjusting method according to one embodiment of the present invention;

5 is a view showing a system configuration of a radio control robot flag game machine according to an embodiment of the present invention.

6 is a view showing the structure of the flag portion of FIG.

7 and 8 illustrate a behavior-based robot control method according to an embodiment of the present invention.

Claims (16)

In the behavior-based robot control device, A physical sensor unit for acquiring image information about the environment and detecting proximity of an opponent robot or an obstacle; A logical sensor unit for outputting external situation information regarding the position and direction of objects in the environment sensed by the physical sensor unit and the risk of collision with an opponent robot or an obstacle; An internal state module for recognizing internal state information of the robot, An external situation evaluation module for evaluating a current external situation using the external situation information; An action set / coordination module for outputting an adjustment result signal by an action set and an adjustment method selected according to an external situation evaluated by the external situation evaluation module and an internal state of the robot recognized by the internal state module; A situation-independent action set / coordination module for outputting an adjustment result signal by a selected situation-independent action set and a situation-independent adjustment method when a collision risk signal is output from the logical sensor unit; A motion controller for generating robot path data according to an adjustment result signal output from the action set / adjustment module or the situation independent action set / adjustment module; And an actuator for driving the robot according to the robot path data generated by the motion controller. The method according to claim 1, The physical sensor unit, A sensor having a physical form, comprising: a CCD camera for acquiring image information about an environment, and an infrared sensor for detecting a proximity of an opponent robot or an obstacle. The method according to claim 1, The logical sensor unit, A virtual sensor based on the physical sensor unit, comprising: a target detection sensor for detecting a position and a direction of an object in an environment, and a collision detection sensor for detecting a collision risk with an opponent robot or an obstacle. Behavior Based Robot Control. The method according to claim 1, Behavior-based robot control apparatus, characterized in that the internal state of the internal state module is set by the user. The method according to claim 1, And a short-term memory for storing external situation information output from the logical sensor unit. The method according to claim 5, And an internal state of the internal state module is changed according to the information extracted from the short-term memory. The method according to claim 1, And an adjustment result signal output from the situation independent action set / adjustment module is changed according to the internal state information of the robot transmitted from the internal state module. As a behavior-based robot control method, (a) outputting external situation information on the position and direction of an object in the environment and on the risk of collision with an opponent robot or an obstacle from the sensor unit; (b) evaluating a current external situation through the external situation information by an external situation evaluation module; (c) recognizing the internal state of the robot by the internal state module; (d) outputting an adjustment result signal by an action set and an adjustment method selected according to the evaluated external situation and the internal state of the robot; (e) robot path data is generated according to the output adjustment result signal, and the robot is driven by the generated robot path data. The method according to claim 8, Step (a) is, Image information about the environment is acquired by the physical sensor and the proximity of the opponent robot or obstacle is detected, and external situation information about the location and direction of objects in the environment and the risk of collision with the opponent robot or obstacle by the logical sensor. Behavior-based robot control method comprising the step of outputting. The method according to claim 9, and (f) outputting an adjustment result signal based on the selected situation independent action set and the situation independent adjustment method when the collision danger signal is output from the logical sensor. The method according to claim 10, Adjusting result signal output in the step (f) is behavior-based robot control method characterized in that it is changed according to the internal state information of the robot transmitted from the internal state module. The method according to claim 10, (g) robot path data is generated according to the adjustment result signal output in step (f), and the robot-based behavior control method further comprises the step of driving the robot by the generated robot path data. The method according to claim 8, In step (c), The internal state of the internal state module is a behavior-based robot control method, characterized in that set by the user. The method according to claim 8, External situation information output in the step (a) is stored in a short-term memory behavior-based robot control method. The method according to claim 14, In step (c), And an internal state of the internal state module is changed according to information extracted from the short-term memory. The method according to claim 8, The adjustment result signal output in the step (d) is a behavior-based robot control method, characterized in that the adjustment result signal by the vector sum of the selected behavior set.

Images