KR101119788B1 - Apparatus for driving mobile robot and method for driving thereof - Google Patents

Abstract

The present invention relates to a device for driving a mobile robot and a method of driving the mobile robot which avoids obstacles during autonomous driving and autonomous driving of the mobile robot, and recognizes the position of the mobile robot and the position of the obstacle, and the size and position of the obstacle. Recognizes at least one obstacle information of the movement direction, and the moving speed, and corrects the movement path of the mobile robot by using the obstacle information.

Description

Apparatus for driving mobile robot and method for driving

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a mobile robot, and more particularly, to a driving apparatus of a mobile robot and a driving method of the mobile robot, which avoid obstacles during autonomous driving and autonomous driving of the mobile robot.

In general, robots have been developed for industrial purposes and used as part of factory automation, or have been used to perform tasks on behalf of humans in environments in which humans cannot work. The robot field has recently been used in the state-of-the-art space development industry and has been evolving, replacing the medical device as well as the human-friendly home robot in recent years.

In addition, with the recent development of mobile robot technology, mobile robots are being used in various forms such as industrial robots, cleaning robots, home robots, educational robots, and medical robots. The autonomous driving method of the mobile robot includes a method of generating a map by the mobile robot itself and driving with reference to the generated map, and a method of randomly driving without making a map such as a cleaning robot.

In the autonomous driving method of the mobile robot, a map is generated and the driving is performed by referring to the generated map, and a person generates a map by manually dragging the mobile robot or moving the robot by itself. Here, the moving distance of the mobile robot is measured using odometry information caused by the wheel rolling of the mobile robot, and a map is generated using the measured moving distance of the mobile robot. In this case, when the wheel of the mobile robot is slipped or caught in an obstacle, the moving distance of the mobile robot may not be accurately measured using the odometry information, and thus, accurate map generation is difficult.

In addition, in the manner of generating the map and driving using the generated map, the mobile robot recognizes the position of the mobile robot and moves autonomously in order to move the robot by referring to the created map after creating the map. Here, the mobile robot recognizes its position using the odometry information. In this case, when the wheel of the mobile robot is slipped or caught in an obstacle, the position of the mobile robot cannot be accurately measured using odometry information, and thus, it is difficult to accurately recognize the location of the mobile robot.

In other words, an error occurs in the rolling of the wheel due to the sliding or locking of the wheel of the mobile robot, and the measurement of the moving distance of the mobile robot or the location of the mobile robot is performed by using incorrect odometry information. It is difficult to generate and recognize the position of the mobile robot, and thus, normal autonomous driving of the mobile robot is difficult.

In addition, a sensor (eg, an infrared or ultrasonic sensor) is used to avoid obstacles during autonomous driving of the mobile robot, and the sensor is attached to the front and side surfaces of the mobile robot. At this time, if an obstacle exists in front of the mobile robot during autonomous driving of the mobile robot, the presence of the obstacle is detected using the sensor, the distance information with the obstacle is recognized using the odometry information, and the obstacle The obstacle is avoided by using the distance information with the. Here, when avoiding the obstacle, it is difficult to recognize the size, the moving direction, the speed, and the like of the obstacle, and it is difficult to apply the mobile robot to the complex real life by simply avoiding the obstacle using only the distance information with the obstacle.

Accordingly, an object of the present invention is to provide a device for driving a mobile robot and a method for driving the mobile robot, which accurately recognizes a position, a moving distance, and an obstacle of the mobile robot to avoid obstacles during autonomous driving and autonomous driving of the mobile robot. .

In addition, another object of the present invention, by using the landmark (landmark) information to accurately recognize the moving distance and location of the mobile robot to generate a map, with reference to the map generated by the mobile robot autonomous driving in the optimal path The present invention provides a driving apparatus for a mobile robot and a driving method for the mobile robot.

In addition, another object of the present invention, the driving device of the mobile robot and the method of driving the mobile robot to efficiently avoid the obstacle by recognizing the size, position, direction of movement, moving speed, etc. of the obstacle during autonomous driving using the sensor. In providing.

The method of the present invention for achieving the above object, the step of recognizing the position of the position and the obstacle of the mobile robot: Recognizing the obstacle information of at least one of the size, position, direction of movement, and movement speed of the obstacle; And modifying a movement path of the mobile robot using the obstacle information.

An apparatus of the present invention for achieving the above objects, the position recognition unit for recognizing the position of the mobile robot; An obstacle recognition unit configured to recognize at least one of the size, position, movement direction, and movement speed of the obstacle using an obstacle sensor in which the light emitting unit and the light receiving unit corresponding to the light emitting unit are alternately layered; And a controller configured to generate a movement path in consideration of the position of the mobile robot and the obstacle, and to move the mobile robot to a destination.

According to the present invention, by accurately recognizing the position and the moving distance of the mobile robot by using the landmark information, it is possible to generate an accurate map and to allow the mobile robot to autonomously travel in an optimal path with reference to the generated map. Accordingly, the present invention can efficiently run autonomously regardless of the external environment such as obstacles. In addition, by using the sensor to recognize the size, location, direction of movement, moving speed, etc. of the obstacle, it is possible to more efficiently avoid the obstacle during autonomous running of the mobile robot. Therefore, the present invention can easily apply a mobile robot to a complex environment such as real life in which various obstacles exist.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. It should be noted that in the following description, only parts necessary for understanding the operation according to the present invention will be described, and descriptions of other parts will be omitted so as not to distract from the gist of the present invention. Then, the driving device of the mobile robot according to the embodiment of the present invention will be described in more detail with reference to FIG. 1.

1 is a view schematically showing a structure of a driving device of a mobile robot according to an embodiment of the present invention.

Referring to FIG. 1, the driving device of the mobile robot may include an obstacle sensor 105 for detecting an obstacle during autonomous driving of the mobile robot, an obstacle recognition unit 110 for recognizing an obstacle detected by the obstacle sensor 105, and a movement. Distance measuring unit 115 for measuring the moving distance of the robot, the position sensor 120 for detecting the position of the mobile robot, the position sensor 120 for recognizing the position of the mobile robot sensed by the position sensor 120 The controller 130 controls autonomous driving and obstacle avoidance of the mobile robot, a storage unit 135 for storing a map for autonomous driving of the mobile robot, a driving unit 140 for moving the mobile robot, and a power source for the mobile robot. It includes a power supply unit 146 for supplying.

The obstacle sensor 105 has a multi-layer structure in which N sensor modules are arranged in M layers, and detects an obstacle where a mobile robot is located by using infrared rays or ultrasonic waves, and detects an obstacle detection signal by an obstacle recognition unit. Transmit to 110. Here, the obstacle sensor 105 will be described in more detail with reference to FIGS. 4 and 5.

The obstacle recognizing unit 110 receives the detection signal of the obstacle from the obstacle sensor 105, and uses the detection signal of the obstacle to detect the size, location, direction of movement, and movement of the obstacle detected by the obstacle sensor 105. Recognize moving speed Here, the obstacle recognition unit 110 calculates the distance between each sensor module arranged in a multi-layer structure of the obstacle sensor 105 and the obstacle, and using the calculated distance, the size, position, direction of movement, And moving speed and the like. The obstacle recognition unit 110 provides the controller 130 with the size, location, movement direction, and movement speed of the obstacle. Here, the obstacle recognition unit 110 will be described in more detail with reference to FIGS. 4 and 5.

The distance measuring unit 115 calculates the rotation amount of the wheel of the mobile robot, measures the moving distance of the mobile robot using the calculated wheel rotation amount, and calculates the moving distance information including the moving distance of the mobile robot. Provided to the controller 130. In this case, the controller 130 mainly uses the information received from the position recognition unit 125 for recognizing the position and the moving distance of the mobile robot, and assists the information received from the distance measuring unit 115.

The position sensor 120 detects a landmark existing on the ceiling of the location where the mobile robot is located, and transmits a landmark detection signal to the position recognition unit 125. Here, the position sensor 120 may have a multilayer structure like the obstacle sensor 105, and may detect the landmark by using infrared rays or ultrasonic waves. Here, the landmark is formed as a different mark indicating a corresponding position for each position is previously displayed on the ceiling of a place (space) for autonomous running of the mobile robot. For example, when the mobile robot runs autonomously in the home, different marks indicating the home's home, living room, and kitchen are displayed on the ceilings of the home, the living room, and the kitchen, respectively. Different symbols representing the position are respectively displayed. Accordingly, a landmark to be used for autonomous driving of the mobile robot is formed on the ceiling of the home.

The position recognition unit 125 receives a landmark detection signal from the position sensor 125 and recognizes the position and the moving distance of the mobile robot using the landmark detection signal. Here, the position recognition unit 125 recognizes the position of the mobile robot by identifying a landmark existing on the ceiling of the place where the mobile robot is located by using the landmark detection signal. In addition, when the mobile robot moves, the location recognition unit 125 changes the landmark identified according to the movement, and recognizes the movement distance of the mobile robot in consideration of the change of the landmark. In addition, the position recognition unit 125 provides the position and the movement distance of the mobile robot to the controller 130.

The controller 130 receives the position, size, movement direction, and movement speed of the obstacle from the obstacle recognition unit 110, and receives the movement distance and the position of the mobile robot from the position recognition unit 125. . The controller 130 drives the mobile robot through the driver 140 using the provided parameters and controls the mobile robot. In addition, the controller 130 may refer to the size and position of the obstacle from the obstacle recognition unit 110 and the moving distance and the position of the mobile robot provided from the position recognition unit 125 when autonomous driving of the mobile robot is performed. Create a map. The controller 130 stores the generated map in the storage 135. Here, the map generation is a pre-operation for autonomous driving of the mobile robot and generates a map in consideration of the moving distance and location of the mobile robot and the size and location of the obstacle.

In addition, the controller 130 searches for the optimal autonomous driving path of the mobile robot in consideration of obstacles on the map, and searches for the optimal autonomous driving path so that the mobile robot autonomously travels on the optimal path during autonomous driving of the mobile robot referring to the map. Set autonomous driving route of autonomous driving route of mobile robot. The controller 130 sets obstacle avoidance to an optimal path on the map in consideration of the position, size, movement direction, and movement speed of the obstacle provided from the obstacle recognition unit 110 during autonomous driving. . Accordingly, the mobile robot avoids obstacles in the optimal path set.

The driving unit 140 moves the mobile robot under the control of the controller 130, and moves the mobile robot by driving a motor by the power supplied from the power supply unit 145. Next, the obstacle sensor 105 in the driving apparatus of the mobile robot according to the embodiment of the present invention will be described in more detail with reference to FIGS. 2 and 3.

2 is a view schematically illustrating a structure of an obstacle sensor in a driving apparatus of a mobile robot according to an embodiment of the present invention, and FIG. 3 is a circuit diagram of an obstacle sensor in a driving apparatus of a mobile robot according to an embodiment of the present invention.

2 and 3, the obstacle sensor 105 has a multilayer structure in which N sensor modules are arranged in M layers, for example, a multilayer structure 210 in which 25 sensor modules are arranged in three layers. Each sensor module includes a light emitting unit 220 and a light receiving unit 230, respectively.

The light emitting unit 220 may include a first resistor unit 305 connected to a first voltage terminal, an output unit 310 connected to the first resistor unit 310 and emitting a signal for detecting a distance of an obstacle, and And a switching unit 315 connecting the output unit 310 to a ground terminal and modulating a signal emitted from the output unit 310. The light receiver 230 includes a second resistor 320 connected to a second voltage terminal, a receiver 325 connected between the second resistor 320 and a ground terminal and receiving a signal reflected from an obstacle. And a filter 330 for filtering the signal received by the receiver 325. Here, the switching unit 315 may be a transistor, and the filter 330 is a band pass filter (BPF) that filters only a signal of a corresponding sensor module from a signal input to the light receiving unit 230. BPF '). The receiver 325 receives the reflected signal when the signal emitted from the output unit 310 of the light emitter 220 is reflected from an obstacle.

Each sensor module detects an obstacle using a signal emitted from the light emitter 220 and a signal received by the light receiver 230, and transmits a detection signal of the obstacle to the obstacle recognition unit 110. The obstacle recognition unit 110 receives the detection signal of the obstacle from each sensor module of the obstacle sensor 105 and uses the detection signal of the obstacle to determine the position, size, movement direction, and moving speed of the obstacle. Recognize.

In more detail, when the light emitter 220 of each sensor module emits a signal to detect the distance of the obstacle, the emitted signal is reflected on the obstacle and the reflected signal is received by the light receiver 230. . Then, each sensor module transmits a detection signal of the obstacle to the obstacle recognition unit 110 according to the signal emission of the light emitting unit 220 and the signal reception of the light receiving unit 230. The obstacle recognition unit 110 calculates a time when the signal emitted from the light emitting unit 220 is reflected by the obstacle and received by the light receiving unit 230, and the distance between each sensor module and the obstacle through the calculated time. Recognize the distance of the obstacle by calculating.

The obstacle recognition unit 110 calculates the distance between each sensor module and the obstacle, and then recognizes the position and size of the obstacle by combining the calculated distances. For example, the distance from the obstacle calculated by the sensor module # 1 is 30, the distance from the obstacle calculated by the sensor modules # 2, # 3, and # 4 is 20, and the sensor module # 5 is calculated. When the distance to one obstacle is 30, the obstacle recognition unit 110 recognizes that the obstacle is located at a distance of 20 from the mobile robot, and between the sensor modules # 2, # 3, and # 4. Recognize the size of obstacles through distance. In addition, when the light emitting unit 220 of each sensor module emits 15 signals per second and the light receiving unit 230 receives the reflected signal, the light emitting unit 220 recognizes the distance, size, and position of the obstacle at 15 frames per second. In each frame, the moving speed and the moving direction of the obstacle are recognized by using the distance between the obstacle, the size, and the position of the obstacle. After recognizing the distance, position, size, moving speed, and moving direction of the obstacle, the obstacle recognizing unit 110 controls obstacle information including the distance, position, size, moving speed, and moving direction of the obstacle. Provided at 130.

Here, when the light emitting unit of the sensor module # 1 in each sensor module emits a signal, the emitted signal is reflected from the obstacle, when the reflected signal is input to all the light receiving unit of each sensor module, obstacle recognition unit 110 It may be difficult to calculate the exact distance between each sensor module and the obstacle, but as each sensor module detects the obstacle in two ways to be described later, the obstacle recognition unit 110 accurately calculates the distance from the obstacle.

First, from the first method, the light emitting units of the respective sensor modules emit signals having different frequencies, and the signals having the different frequencies are reflected from obstacles and input to all light receiving units of the respective sensor modules. Here, the light receiving unit includes a BPF for filtering only a signal having a frequency corresponding to each sensor module. Therefore, the light receiving unit of each sensor module receiving the signals having different frequencies filters the signals of the corresponding sense modules from the signals having different frequencies using BPF.

That is, each of the sense modules emits signals having different frequencies to detect obstacles, and filters and receives only signals having frequencies corresponding to them by using BPF from signals reflected from the obstacles. As the sense module emits and receives only a signal corresponding to the sensor, the obstacle recognition unit 110 can accurately calculate the distance between each sensor module and the obstacle, and the distance between the obstacle calculated as described above. Recognize the position and size of the obstacle by using a combination, and also recognize the movement speed and the direction of movement of the obstacle using the distance, size, and position of the obstacle in each frame.

In the second method, the light emitting unit of each sensor module sequentially emits a signal for each sensor module, and the emitted signal is reflected from an obstacle and received by the light receiving unit of each sensor module. Here, the light receiving unit of each sensor module receives the reflected signal by sequentially turning on the power for each sensor module. In other words, when the light emitting unit of the sensor module # 1 is turned on and emits a signal, only the light receiving unit of the sensor module # 1 is turned on, and receives only the light receiving unit of the sensor module # 1 receiving a signal reflected from an obstacle. do. In addition, the light emitting unit and the light receiving unit of sensor module # 2, the light emitting unit and the light receiving unit of sensor module # 3 are sequentially turned on, and in each sensor module as power is sequentially turned on to the light emitting unit and the light receiving unit of the last sensor module. The signal is emitted and received sequentially for each sensor module. Here, the power of the light emitting unit and the light receiving unit of each sensor module is only the power of the light emitting unit and the light receiving unit of the corresponding sensor module, the power of the light emitting unit and the light receiving unit of the remaining sensor module is off (off), the light emitting unit of each sensor module And power on / off of the light receiving unit is controlled from the controller 130.

As the emission and reception of signals in each sensor module are sequentially performed for each sensor module, the obstacle recognition unit 110 may accurately calculate the distance between each sensor module and the obstacle. In addition, the obstacle recognition unit 110 recognizes the position and size of the obstacle by combining the distance with the obstacle calculated as described above, and further, by using the distance between the obstacle, the size, and the position in each frame. Recognize the speed and direction of movement of obstacles. Next, the pre-operation process of generating a map referenced during autonomous driving of the mobile robot will be described in detail with reference to FIG. 4.

4 is a diagram schematically illustrating a map generation process of a driving device of a mobile robot according to an exemplary embodiment of the present invention.

Referring to FIG. 4, in step S410, a landmark existing on a ceiling of a location where a mobile robot is located is detected in advance so that the mobile robot recognizes a path to be moved, and the location and direction of the detected landmark are calculated, and the calculation is performed. The location and the movement distance of the mobile robot are recognized using the location and direction of the landmark.

In operation S420, a path for moving the mobile robot using the recognized position of the mobile robot is calculated as a distance, and accordingly, a moving distance of the mobile robot is calculated. Here, the moving distance of the mobile robot may calculate the amount of rotation of the wheel of the mobile robot, measure the moving distance of the mobile robot using the calculated amount of wheel rotation, and calculate the calculated distance using the measured moving distance of the mobile robot. have.

In operation S430, the distance of the obstacle is calculated using a sensor module having a multilayer structure, and the size and location of the obstacle are recognized using the distance of the obstacle. Then, in step S440 generates a map using the location and the moving distance of the mobile robot, and the size and location of the obstacle. Then, when a map is generated by a pre-operation for autonomous driving of the mobile robot with reference to FIG. 5, obstacle avoidance during autonomous driving and autonomous driving of the mobile robot with reference to the generated map will be described.

5 is a diagram schematically illustrating an autonomous driving operation process of a mobile robot according to an exemplary embodiment of the present invention.

Referring to FIG. 5, in step S510, the mobile robot receives a destination to be moved by autonomous driving. In operation S520, the mobile robot automatically sets an optimal route to the input destination by referring to the previously generated map as described with reference to FIG. 4. Next, in step S530 it detects the landmark on the ceiling of the location where the mobile robot is located, recognizes the location of the mobile robot by using the detected position of the landmark, and starts autonomous driving in the optimal path set do.

In step S540, the existence of an obstacle is checked while autonomous driving is performed in the optimal path. In this case, the existence of the obstacle is determined by the light emitting unit of each sensor module emitting a signal for detecting the obstacle as described above, and the light receiving unit of each sensor module receives a signal reflected from the obstacle.

If there is an obstacle as a result of the check in step S540, in step S550, the distance to the obstacle is calculated, and the position and size of the obstacle are recognized by combining the calculated distances. Using the difference between the vehicle to recognize the moving speed and direction of the obstacle. In step S560, the obstacle avoidance from the set optimal path to the optimal path on the generated map is set in consideration of the position, the size, the moving direction, the moving speed, and the like of the obstacle.

Then, the identified obstacle is avoided through the obstacle avoidance set in the step S570. In operation S580, the autonomous vehicle arrives at the input destination by autonomous driving in an optimal path, and detects a landmark existing on the ceiling of the location where the mobile robot arrives, and uses the location of the detected landmark to locate the mobile robot. Recognize.

Meanwhile, in the detailed description of the present invention, specific embodiments have been described, but various modifications are possible without departing from the scope of the present invention. Therefore, the scope of the present invention should not be limited to the described embodiments, but should be defined not only by the scope of the following claims, but also by the equivalents of the claims.

1 is a view schematically showing a structure of a driving device of a mobile robot according to an embodiment of the present invention.

2 is a view schematically showing the structure of an obstacle sensor in a driving device of a mobile robot according to an embodiment of the present invention.

3 is a circuit diagram of an obstacle sensor in a driving device of a mobile robot according to an embodiment of the present invention.

4 is a view schematically illustrating a map generation process of a driving device of a mobile robot according to an embodiment of the present invention.

5 is a view schematically showing a process of autonomous driving of a mobile robot according to an embodiment of the present invention.

Claims (19)

In the driving device of the mobile robot, A position recognition unit recognizing a position of the mobile robot; An obstacle recognition unit configured to recognize at least one of the size, position, movement direction, and movement speed of the obstacle using an obstacle sensor in which the light emitting unit and the light receiving unit corresponding to the light emitting unit are alternately layered; And A controller configured to generate a movement path in consideration of the position of the mobile robot and the obstacle, and move the mobile robot to a destination; The light emitting unit includes a first resistor unit connected to a first voltage terminal, an output unit connected to the first resistor unit and outputting a transmission signal, and a switching unit connecting the output unit to a ground terminal and modulating the transmission signal. Driving device for a mobile robot, characterized in that. The method of claim 1, The light emitting unit includes a plurality of light emitting sensors per layer, and each light emitting sensor outputs an outgoing signal having a different frequency. delete The method of claim 1, The light receiving unit includes a plurality of light receiving sensors for each floor, and each light receiving sensor is a driving device of a mobile robot, characterized in that for receiving by filtering the transmission signal output from the light emitting unit reflected from the obstacle for each frequency. The method of claim 1, The light receiving unit may include a first resistor unit connected to a first voltage terminal, a receiver connected between the first resistor unit and a ground terminal to receive an outgoing signal reflected from the obstacle, and output from the light emitting unit and reflected from the obstacle. And a band pass filter (BPF) for filtering the outgoing signal for each frequency. The method of claim 1, The light emitting unit includes a plurality of light emitting sensors per layer, and the plurality of light emitting sensors sequentially drive the mobile robot, characterized in that the output signal. The method of claim 1, The light receiving unit includes a plurality of light receiving sensors for each floor, and the plurality of light receiving sensors are sequentially driven by the light emitting unit and receive the transmission signal reflected from the obstacle. The method of claim 1, The obstacle recognizing unit recognizes the position of the obstacle by using the time when the outgoing signal output from the light emitting unit is reflected by the obstacle received by the light receiving unit. The method of claim 8, The obstacle recognizing unit recognizes the position of the obstacle for each frame, and using the position difference for each frame of the obstacle to recognize the movement speed and the direction of movement of the obstacle robot. The method of claim 1, The obstacle recognizing unit recognizes the size of the obstacle using the number of light receiving sensors reflected from the obstacle received by the light receiving unit. The method of claim 1, When the obstacle is recognized during the movement of the mobile robot, the controller modifies the movement path in consideration of at least one of the size, position, movement direction, and movement trajectory of the obstacle, and moves the movement to the destination. Driving device of the robot. The method of claim 1, And the location recognizing unit recognizes a location of the mobile robot by using a sensor that detects a landmark indicating a location and a movement path of the mobile robot. In the driving method of a mobile robot, Recognizing the position of the mobile robot and the position of the obstacle: Recognizing at least one obstacle information among a size, a position, a moving direction, and a moving speed of the obstacle using an obstacle sensor in which a light emitting unit and a light receiving unit corresponding to the light emitting unit alternately form a layer; The light emitting part includes a first resistor part connected to a first voltage terminal, an output part connected to the first resistor part and outputting an outgoing signal, and a switching part connecting the output part to a ground terminal and modulating the outgoing signal. -And And modifying a movement path of the mobile robot using the obstacle information. delete The method of claim 13, The light emitting unit outputs an outgoing signal having a different frequency from a plurality of light emitting sensors arranged in layers, and the light receiving unit filters out the outgoing signal reflected from the obstacle by a plurality of light receiving sensors corresponding to the plurality of light emitting sensors by frequency. A method of driving a mobile robot, characterized in that receiving. The method of claim 13, The light emitting unit sequentially outputs a transmission signal by a plurality of light emitting sensors configured for each layer, and the light receiving unit sequentially receives a transmission signal reflected from the obstacle by a plurality of light receiving sensors corresponding to the plurality of light emitting sensors. Method of driving a mobile robot. The method of claim 13, Recognizing the obstacle information, the driving method of the mobile robot, characterized in that for recognizing the position of the obstacle using the time that the outgoing signal output from the light emitting unit is reflected from the obstacle received by the light receiving unit. The method of claim 13, Recognizing the obstacle information, the driving method of the mobile robot, characterized in that for recognizing the size of the obstacle using the number of light receiving sensors reflected from the obstacle received by the light receiving unit. The method of claim 13, Recognizing the obstacle information, the location of the obstacle for each frame, the method of driving the mobile robot, characterized in that for recognizing the moving speed and the direction of movement of the obstacle using the position difference of each frame of the obstacle. .

Images