WO2017041225A1

WO2017041225A1 - Robot following method

Info

Publication number: WO2017041225A1
Application number: PCT/CN2015/089127
Authority: WO
Inventors: 覃国秘; 谭和华; 罗辉; 钟志威
Original assignee: 深圳市赛亿科技开发有限公司
Priority date: 2015-09-08
Filing date: 2015-09-08
Publication date: 2017-03-16
Also published as: CN107073711A

Abstract

A robot following method, comprising: recognizing a tracked object; monitoring the intensity of a Bluetooth signal received by two Bluetooth receiving modules (132) which are arranged on the robot at an interval of L meters; according to the monitored intensity of the Bluetooth signal, respectively calculating distances L1 and L2 between the two Bluetooth receiving modules (132) and a Bluetooth transmitting module (131); acquiring foremost two groups of L1 and L2, and according to L, L1 and L2, calculating the speed and direction of the tracked object; according to the calculated speed and direction of the tracked object, adjusting the speed and direction of the robot; receiving a detection range of a sensor in each group and according to a pre-set rule, acquiring a minimum value of each group to be used as a determined detection range of this group; according to the latest determined detection range, determining the distance and direction of an obstacle; and according to the speed of the robot and changes in time and distance of the detection range sent by a sensor module twice, determining a position of an obstacle and bypassing the obstacle. The robot following method has the advantages of being simple in use, low in cost and easy to popularize in a wide range.

Description

Robot following method

Technical field

The present invention relates to the field of robot technology, and in particular, to a robot following method.

Background technique

With the development of society and the advancement of technology, robots occupy more and more positions in production and life. Due to the widespread use of microprocessor controllers and sensors, the functionality and precision of modern robots are becoming more sophisticated. The follow-up function of the robot plays an important role in the robot follow-up method, and it plays an important role in factory production, hospitals, and shopping malls.

Most of the current follow-up robots use a computer to specify the objects to be followed by live video, and use the computer as a basis to avoid obstacles. This kind of robot is complicated to use and high in cost, which is not conducive to popularization and application.

Summary of the invention

It is an object of the present invention to provide a robot following method that is easy to use and low in cost, and has a function of automatically following a tracking object and automatically avoiding an obstacle.

In order to solve the above technical problems, the invention adopts the technical solution described below. A robot following method, the robot includes a Bluetooth module and a sensor module, the Bluetooth module includes two Bluetooth receiving modules and a Bluetooth transmitting module, the sensor module includes a plurality of first sensors and a plurality of second sensors, the plurality of The first sensor and the plurality of second sensors form at least one group, and the method includes: identifying a tracking object, the tracking object is configured with the Bluetooth transmitting module; and monitoring two Bluetooth receiving modules on the robot with an interval of L meters to receive the Bluetooth The strength of the signal; calculate the distances L1 and L2 between the two Bluetooth receiving modules and the Bluetooth transmitting module according to the detected Bluetooth signal strength; obtain the latest two groups of L1 and L2, and calculate the tracking object according to L, L1 and L2 Speed and direction; Calculating the speed and direction of the tracking object to adjust the speed and direction of the robot; receiving the detection distances of the first sensor and the second sensor in each group; acquiring the minimum value of each group according to a predetermined rule as the detection distance determined by the group; The two detection distances and the speed and direction of the robot itself to determine the distance and direction of the obstacle; according to the speed of the robot and the change of the time and distance of the detection distance sent by the two sensor modules before and after, determine the position of the obstacle and bypass the obstacle The object is tracked.

Preferably, the calculating the distances L1 and L2 between the two Bluetooth receiving modules and the tracking object according to the monitored Bluetooth signal strength respectively includes: collecting a current RSSI value of the Bluetooth signal strength, and performing current RSSI values according to a preset filtering algorithm. Processing, obtaining the current RSSI base value; using the offline BP neural network algorithm to process the current RSSI base value, and obtaining the distances L1 and L2 between the Bluetooth transmitting module and the two Bluetooth receiving modules.

Preferably, the acquiring the latest two groups L1 and L2 and calculating the speed and direction of the tracking object according to L, L1 and L2, and adjusting the speed and direction of the robot according to the calculated speed and direction of the tracking object comprises: obtaining the speed of the robot; According to the distance between the Bluetooth transmitting module and the two Bluetooth receiving modules measured twice before and after, the speed of the tracking object is obtained, and the speed of the robot is adjusted according to the speed of the obtained tracking object, which is consistent with the speed of the tracking object; The distance between the Bluetooth transmitting module and the two Bluetooth receiving modules calculates the steering angle of the tracking object, and adjusts the direction of the robot to be consistent with the direction of the tracking object according to the obtained steering angle of the tracking object.

Preferably, the sensor module includes a plurality of first sensors and a plurality of second sensors, and the plurality of first sensors and the plurality of second sensors form at least one group: the sensor module is five preset on a semicircular arc of the robot a first sensor and four second sensors, and five first sensors and four second sensors are arranged in an arc length interval, and the adjacent two first sensors and one second sensor are set as a group Divided into three groups.

Preferably, the first sensor is an ultrasonic sensor and the second sensor is an infrared sensor.

Preferably, the obtaining the minimum value of each group according to a predetermined rule as the detection distance determined by the group includes: when the distance of detecting the obstacle is within the first preset range, taking the detection distance of the second sensor as the group The determined detection distance; when the distance of the obstacle is detected to be within the second preset range, The detection distance of a sensor is used as the detection distance determined by the group; when the distance of the obstacle is detected within the third preset range, the detection distance of the first sensor and the detection distance of the second sensor are merged to obtain the group. Determined detection distance.

Preferably, the combining the detection distance of the first sensor and the detection distance of the second sensor to obtain the determined detection distance of the group, the fusion method is: d=a×x+b×y, where d is The distance of the obstacle, x is the detection distance of the first sensor, y is the detection distance of the second sensor, and the values of a and b are determined by the values of x and y. When x is in the range of 0.6 to 0.8, the value of a In the range of 0.5 to 1, the value of a is proportional to the value of x; when y is in the range of 0.4 to 0.6, the value of y varies from 0.5 to 1, y and b is in a negative proportional relationship; when x>0.6, y<0.6, the values of a and b are both 0.5.

Preferably, the determining the position of the obstacle and tracking the obstacle according to the change of the time and distance of the speed of the robot and the sending of the detection distance by the two sensor modules before and after the method comprises: determining the obstacle according to the detection distance determined twice before and after The distance and direction, the movement state of the obstacle, the speed of movement and the size of the obstacle.

Preferably, the robot following method further comprises: pre-storing a plurality of specified gestures of the tracking object, and one-to-one correspondence between the specified gestures and the distance between the robot and the tracking object; identifying a plurality of specified gestures of the tracking object, according to the identified Specify the gesture to adjust the robot to the corresponding distance.

Preferably, the one-to-one correspondence between the plurality of specified gestures and the distance between the robot and the tracking object includes: the plurality of specified gestures include beckoning, waving, and raising a hand, and adjusting the robot to the following object when the tracking object is beckoning The specified distance is closer. When the tracking object swings, the adjustment robot retreats to a specified distance farther from the following object. When the tracking object raises the hand, the robot stops at its original position.

The beneficial technical effect of the present invention is that the robot following method establishes a connection by a Bluetooth transmitting module disposed on the tracking object and two Bluetooth receiving modules disposed on the robot, thereby calculating a distance and a direction of the tracking object, thereby implementing the tracking object. Automatically follow. At the same time, the distance and direction of the obstacle are determined by the detection distance determined twice before and after, and the position of the obstacle is determined according to the speed of the robot and the time and distance of the detection distance sent by the two sensor modules before and after, thereby achieving the obstacle avoidance. . The robot following method has the functions of automatically following the tracking object and automatically avoiding the obstacle, and has the advantages of simple use, convenient operation, low cost, high intelligence, and easy wide-scale promotion.

DRAWINGS

1 is a structural view of a robot provided by a preferred embodiment.

2 is a distribution diagram of the sensor module of FIG. 1.

FIG. 3 is a flow chart of a robot following method according to a preferred embodiment.

detailed description

The invention will be further described below in conjunction with the drawings and embodiments in order to provide a further understanding of the invention.

1 is a structural view of a robot provided by a preferred embodiment. The robot 1 includes a robot body 11, a microcontroller 12, a Bluetooth module 13, a drive module 14, and a sensor module 15. The Bluetooth module 13, the drive module 14, and the sensor module 15 are electrically connected to the microcontroller 12, respectively. The Bluetooth module 13 includes a Bluetooth transmitting module 131 disposed on the tracking object and two Bluetooth receiving modules 132 disposed on the robot body 11 for determining the distance and direction between the robot body 11 and the tracking object. The sensor module 15 is used to detect obstacles. The microcontroller 12 receives the feedback information of the Bluetooth module 13 and the sensor module 15 to control the driving module 15 to drive the robot body 11 to operate, thereby implementing automatic tracking of the tracking object.

Preferably, in the embodiment, the robot body 11 further includes a control circuit board and a monitoring device 18 disposed on the control circuit board. The microcontroller 12 and the monitoring device 18 are electrically connected to the control circuit board, respectively. The monitoring device 18 is configured to monitor the power of the Bluetooth receiving device 132 to receive the signal, and thereby determine the distance between the robot body 11 and the tracking object.

Referring to FIG. 2, the distribution diagram of the sensor module 15 is such that five ultrasonic sensors 151 and four infrared sensors 152 are equally spaced apart on the semicircular arc to form two adjacent ultrasonic sensors 151 and the middle. The infrared sensor 152 is a group of three groups to make the sensor module 15 It can detect obstacles within 180 degrees of the front of the robot body 11, thereby increasing the detection range, avoiding obstacles as much as possible, and achieving safe and collision-free tracking.

In this embodiment, the driving module 14 includes four rollers and two driving mechanisms. Preferably, the drive mechanism employs a servo motor.

Preferably, the camera body 16 is further disposed on the robot body 11 , and the camera module 16 is electrically connected to the microcontroller 12 . Preferably, in this embodiment, the storage module 11 and the identification module 19 are further disposed on the circuit control board. The storage module 17 is configured to save a preset gesture, and the identification module 19 is configured to recognize according to the gesture captured by the camera module 16 and determine the following action of the robot body 11 accordingly. The gesture image analysis program is stored in advance in the microprocessor controller 12, and the set gesture is saved to the storage module 17. Such gestures as preset may include three gestures of beckoning, waving, and raising a hand. After beckoning, the robot will walk closer to the tracking object; instead, the robot will retreat to a farther distance from the tracking object; after raising the hand, the robot will stop at its original place. Of course, the solution is not limited to these three gestures, and the user can set different types of gestures according to requirements.

Referring to FIG. 3, a flow chart of a robot following method is provided by the preferred embodiment. The robot tracking method 10 includes:

101: Identify a tracking object, where the tracking object is configured with a Bluetooth transmitting module;

102: Monitor the strength of the Bluetooth signal received by the two Bluetooth receiving modules set by the interval L meter on the robot;

103: Calculate distances L1 and L2 between two Bluetooth receiving modules and a Bluetooth transmitting module according to the detected Bluetooth signal strength;

104: Obtain the latest two groups of L1 and L2, and calculate the speed and direction of the tracking object according to L, L1 and L2;

105: Adjusting the speed and direction of the robot according to the calculated speed and direction of the tracking object;

106: receiving a detection distance of the first sensor and the second sensor in each group;

107: Obtain a minimum value of each group according to a predetermined rule as a detection distance determined by the group;

108: Determine the distance and direction of the obstacle according to the latest two detection distances and the speed and direction of the robot itself;

109: According to the speed of the robot and the change of the time and distance of the detection distance sent by the two sensor modules before and after, the position of the obstacle is judged and the obstacle is tracked.

Specifically, 101: identifying a tracking object, where the tracking object is configured with a Bluetooth transmitting module, including:

The tracking object wears a Bluetooth transmitting module and establishes a connection with two Bluetooth receiving modules disposed on the robot. Tracking objects are not limited to people, but also other movable objects. In this embodiment, the tracking object is set to be a person, that is, a user. In use, the user wears a Bluetooth transmitting module, and two Bluetooth receiving modules are arranged on the robot at intervals to realize a connection between the robot and the tracking object through a connection between the Bluetooth transmitting module and the Bluetooth receiving module.

102: Calculate the distances L1 and L2 between the two Bluetooth receiving modules and the tracking object according to the detected Bluetooth signal strength respectively, including:

Specifically, in the embodiment, the robot is provided with a control circuit board and a monitoring device disposed on the control circuit board. The monitoring device is configured to monitor the power of the signal received by the Bluetooth receiving module, and thereby determine the distance between the robot and the tracking object. Specifically, calculating the distances L1 and L2 between the two Bluetooth receiving modules and the tracking object according to the monitored Bluetooth signal strength respectively includes:

Collecting the current RSSI value of the Bluetooth signal strength, and processing the current RSSI value according to a preset filtering algorithm to obtain a current RSSI base value;

The offline BP neural network algorithm is used to process the current RSSI base value, and the distances L1 and L2 between the Bluetooth transmitting module and the two Bluetooth receiving modules are obtained.

When the robot is working, the monitoring device measures the distances L1 and L2 between the robot body and the user by monitoring the signal power received by the two Bluetooth receiving devices. In this embodiment, the distance is measured by a filtering method based on a RSSI (Received Signal Strength Indicator) wireless positioning method and a combination of hopping filtering and Gaussian filtering. In the received current RSSI value, the maximum value and the minimum value are removed, and then the current RSSI base value is obtained by weighted averaging to reduce the error of the RSSI value due to interference. The offline BP (Back Propagation) neural network algorithm is used to calculate the distance between the Bluetooth transmitting module and the two Bluetooth receiving modules by using the obtained current RSSI base value. The propagation formula of Bluetooth signal in free space is:

Where d is the distance between two points, p is the power of attenuation, k is the random variable factor, and α is the signal attenuation factor. Due to the complex use environment, k and α have great uncertainty. The nonlinear continuous mapping function of BP neural network algorithm is used to perform offline measurement simulation. BP neural network consists of input layer, intermediate hidden layer and output layer. Under the current RSSI base value and distance value obtained by actual experiments, the hidden layer is respectively When the number of neurons is a, b, c, d..., the simulation is performed, and then the simulated model distance and the actual error curve are selected, and the number of neurons with the smallest error is selected as the input parameter to calculate the Bluetooth transmitting module and The distance between two Bluetooth receiving modules, L1 and L2.

104: Obtain the latest two groups of L1 and L2, and calculate the speed and direction of the tracking object according to L, L1 and L2, and 105: adjust the speed and direction of the robot according to the calculated speed and direction of the tracking object, including:

Get the speed of the robot;

According to the distance between the Bluetooth transmitting module and the two Bluetooth receiving modules measured twice before and after, the speed of the tracking object is obtained, and the speed of the robot is adjusted according to the speed of the obtained tracking object to be consistent with the speed of the tracking object;

The steering angle of the tracking object is calculated according to the distance between the Bluetooth transmitting module and the two Bluetooth receiving modules measured twice before and after, and the direction of the robot is adjusted according to the steering angle of the obtained tracking object.

Specifically, the driving mechanism of the robot employs a servo motor. The servo motor controls the speed of the algorithm by PID (proportion, integration, differentiation) to control the speed of the robot. Since the servo motor has a built-in Hall sensor, when the servo motor rotates, the Hall sensor emits several sets of signals every time the servo motor rotates, and one of these signals is selected as the test basis, and two adjacent ones are recorded. The time between the pulses, and then the diameter of the roller divided by the pulse time, can calculate the current speed of the roller, and can control the speed of the servo motor through the PID control algorithm, according to which the speed and position adjustment accuracy of the servo motor is very high. Accurate, so that the robot has good follow-up performance. In this embodiment, the drive mechanism includes four rollers. Four wheels It is divided into two pairs of front and rear. Each rear wheel acts as a drive wheel, which is connected to a drive mechanism. The front wheel acts as a driven wheel, which can rotate in any direction. The two Bluetooth receiving modules are set at intervals of L meters. According to the RSSI measurement method, the distances between the Bluetooth transmitting modules on the user and the two Bluetooth receiving modules are respectively L1 and L2, and the distance and direction of the tracking object are calculated by L, L1 and L2. To locate the user's location. Then, the electronic differential steering mode is used for steering, the acceleration between the two rear wheels is performed by two servo motors, and then the respective speeds of the two motors are controlled according to the measured speed and the angle of the required steering, so that the two drives The speed of the wheel is different, and the purpose of steering is achieved, so that the speed and direction of the robot are adjusted according to the obtained speed and direction of the tracking object, and the automatic tracking function of the robot is realized. At the same time, the distance and the steering of the tracking object are measured by the Bluetooth transmitting module and the two Bluetooth receiving modules. Compared with the prior art, the cost is not only reduced, but also the product promotion and practicability are increased.

Specifically, the sensor module includes a plurality of first sensors and a plurality of second sensors, and the plurality of first sensors and the plurality of second sensors form at least one group:

The sensor module is five first sensors and four second sensors preset on the semicircular arc of the robot, and five first sensors and four second sensors are arranged at an arc length interval, and the adjacent two A sensor and a second sensor are arranged in a group and are divided into three groups. Preferably, the first sensor is an ultrasonic sensor and the second sensor is an infrared sensor.

107: Obtain a minimum value of each group according to a predetermined rule as the detection distance determined by the group includes:

When the distance of the obstacle is detected to be within the first preset range, the detection distance of the second sensor is taken as the detection distance determined by the group;

When the distance of the obstacle is detected to be within the second preset range, the detection distance of the first sensor is taken as the detection distance determined by the group;

When the distance of the obstacle is detected within the third preset range, the detection distance of the super first sensor and the detection distance of the second sensor are merged to obtain the determined detection distance of the group.

In order to realize the function of the robot to automatically avoid obstacles, in the embodiment, the sensor module includes five ultrasonic sensors and four infrared sensors, and five ultrasonic sensors and four infrared sensors are arc-length on a semi-arc. Interval installation, forming two adjacent ultrasonic sensors and an intermediate infrared sensor as a group of three groups, so that the sensor module can detect within 180 degrees of the front of the robot Obstacles, thereby increasing the detection range, avoiding obstacles as much as possible, and achieving safe and collision-free tracking. Since the ultrasonic sensor has a relatively accurate detection range of 0.8 to 7 m, and the infrared sensor has a relatively accurate detection range of 0.4 m or less, the sensor module on the robot is divided into three segments. The first segment is measured separately for four infrared sensors, and the measured distance ranges from 0 to 0.4 m. The second segment is the distance between the five ultrasonic sensors and the four infrared sensors measured in the range of 0.4 to 0.8 m, and the third segment is five. The ultrasonic sensors are measured separately and the measured distance ranges from 0.8 to 7 m. When collecting data, the distances measured by the three sensors of each group are fused, and the smaller one is taken as the input data of the group.

Specifically, the detection data of each group of ultrasonic sensors and the detection data of the infrared sensor are fused to obtain input data of the group, and the fusion method is: d=a×x+b×y, where d is an obstacle Distance, x is the detection data of the ultrasonic sensor, y is the detection data of the infrared sensor, and the values of a and b are determined by the values of x and y. When x is in the range of 0.6 to 0.8, the value of a is 0.5 to 1. Within the range, the value of a is proportional to the value of x; when y is in the range of 0.4 to 0.6, the value of y varies from 0.5 to 1, and y is negatively proportional to b. When x>0.6 and y<0.6, the values of a and b are both 0.5. Through this fusion method, it is possible to better compensate for the problem that the ultrasonic sensor is not accurately detected at a close distance and the infrared sensor is not accurately detected at a long distance.

During the tracking process, the ultrasonic sensor and the infrared sensor on the robot detect the surrounding obstacles in real time and send the detection results to the microcontroller. The microcontroller can know which direction the obstacle is in according to the changes of the two detection data before and after. According to the distance detected by the multiple sensors, it can be known which sensors are between the detection angles of the sensors, and according to the distance detected by the sensor, this angle multiplied by twice the distance is the size of the obstacle. The total speed is obtained by dividing the difference between the distances detected before and after, and dividing by the interval between transmissions. Finally, by subtracting the speed of the robot from the total speed, the moving speed of the obstacle can be obtained. Therefore, the sensor module can obtain the position and size of the obstacle, the static or moving, and the moving speed by transmitting the data detected in real time to the microcontroller, and then adjust the motion state according to the structure and size of the structure itself, thereby bypassing. Obstacle tracking. The comprehensive and reliable detection of obstacles is realized in turn, thereby avoiding direct collision between the robot and the obstacle. In addition, since the sensor module is inexpensive to use, and multiple detections are formed by using multiple sensors, the reliability of detection is improved while reducing costs.

Preferably, the robot following method 10 further includes:

Pre-storing a number of specified gestures of the tracking object, and one-to-one correspondence between the specified gestures and the distance between the robot and the tracking object; identifying a number of specified gestures of the tracking object, and adjusting the robot to the corresponding distance according to the identified specified gesture.

Specifically, the one-to-one correspondence between the specified gestures and the distance between the robot and the tracking object includes: the plurality of specified gestures include waving, waving, and raising the hand.

When the tracking object is beckoning, adjust the robot to advance to a specified distance closer to the object to be followed.

When the tracking object swings, adjust the robot to retreat to a specified distance further from the object to be followed.

When the tracking object raises the hand, the robot stops at its original position.

Preferably, a camera module is further disposed on the robot. A number of specified gesture analysis programs are stored in advance in the microprocessor controller, and a number of specified gestures are in one-to-one correspondence with the distance between the robot and the tracking object. Such gestures as preset may include three gestures of beckoning, waving, and raising a hand. After beckoning, the robot will walk closer to the tracking object; instead, the robot will retreat to a farther distance from the tracking object; after raising the hand, the robot will stop at its original place. Of course, the solution is not limited to these three gestures, and the user can set different types of gestures according to requirements.

When the robot is tracking, the position of the tracking object is acquired by the Bluetooth receiving module to adjust the direction of the camera module, so that the camera module always faces the tracking object. A program for judging the distance between the robot and the tracking object according to different gestures is pre-stored in the microcontroller. After the camera module sends the captured gesture of the tracking object to the microcontroller, the microcontroller determines the distance that the robot should keep with the tracking object by recognizing the gesture acquired by the module, and then adjusts its own motion posture to achieve This distance is within range.

When the robot is working, the tracking object is first found through the Bluetooth receiving module and the distance and direction of the tracking object are determined. The Bluetooth receiving module and the sensor module send the monitored data to the micro-controller every 100 ms. In the case of normal tracking, the distance between the robot and the tracking object ranges from 0.5 to 1 meter, and the robot calculates the tracking object based on the detected speed of itself and the distance between two adjacent detected objects. The speed, and adjust your own speed is consistent with the tracking object. When the robot detects the change of the direction of the tracking object, it will automatically adjust the two servo motors. The speed makes the two rear wheels differential to achieve steering tracking. During the tracking process, the sensor module on the robot detects the surrounding obstacles in real time and sends the detection results to the microcontroller. The microprocessor determines the position and size of the obstacle, the static or moving, and the moving speed through data analysis. Then, according to the structure and size of the structure itself, adjust the state of motion to track the obstacles. Compared with the prior art, the robot of the invention has the advantages of simple operation, accurate precision and relatively low cost, high intelligence degree, can be used in a complicated environment, and is easy to be widely promoted.

The above description is only a preferred embodiment of the invention, and is not intended to limit the invention in any way. A person skilled in the art can make various equivalent modifications and improvements to the above-described embodiments, and equivalent changes or modifications made within the scope of the claims should fall within the scope of the invention.

Claims

A robot following method, the robot includes a Bluetooth module and a sensor module, the Bluetooth module includes two Bluetooth receiving modules and a Bluetooth transmitting module, the sensor module includes a plurality of first sensors and a plurality of second sensors, the plurality of The first sensor and the plurality of second sensors form at least one group, and the method includes:

Identifying a tracking object, the tracking object being configured with the Bluetooth transmitting module;

Monitoring the strength of the Bluetooth signal received by the two Bluetooth receiving modules set on the robot at intervals of L meters;

Calculating the distances L1 and L2 between the two Bluetooth receiving modules and the Bluetooth transmitting module according to the monitored Bluetooth signal strength;

Obtain the latest two sets of L1 and L2, and calculate the speed and direction of the tracking object according to L, L1 and L2;

Adjust the speed and direction of the robot based on the calculated speed and direction of the tracked object;

Receiving a detection distance of the first sensor and the second sensor in each group;

Obtaining a minimum value of each group as a detection distance determined by the group according to a predetermined rule;

Determine the distance and direction of the obstacle based on the latest two detection distances and the speed and direction of the robot itself;

According to the speed of the robot and the change of the time and distance of the detection distance sent by the two sensor modules before and after, the position of the obstacle is judged and the obstacle is tracked.
The robot following method according to claim 1, wherein the calculating the distances L1 and L2 between the two Bluetooth receiving modules and the tracking object according to the monitored Bluetooth signal strength respectively comprises:

Collecting the current RSSI value of the Bluetooth signal strength, and processing the current RSSI value according to a preset filtering algorithm to obtain a current RSSI base value;

The offline BP neural network algorithm is used to process the current RSSI base value, and the distances L1 and L2 between the Bluetooth transmitting module and the two Bluetooth receiving modules are obtained.
The robot following method according to claim 1, wherein said acquiring the latest two groups of L1 and L2, and calculating the speed and direction of the tracking object based on L, L1 and L2, according to the calculated speed and direction of the tracking object Adjusting the speed and direction of the robot includes:

Get the speed of the robot;

According to the distance between the Bluetooth transmitting module and the two Bluetooth receiving modules measured twice before and after, the speed of the tracking object is obtained, and the speed of the robot is adjusted according to the speed of the obtained tracking object to be consistent with the speed of the tracking object;

The steering angle of the tracking object is calculated according to the distance between the Bluetooth transmitting module and the two Bluetooth receiving modules measured twice before and after, and the direction of the robot is adjusted according to the steering angle of the obtained tracking object.
The robot following method according to claim 1, wherein the sensor module comprises a plurality of first sensors and a plurality of second sensors, and the plurality of first sensors and the plurality of second sensors form at least one group:

The sensor module is five first sensors and four second sensors preset on the semicircular arc of the robot, and five first sensors and four second sensors are arranged at an arc length, and the adjacent two are The first sensor and the second sensor are arranged in a group and are divided into three groups.
The robot following method according to claim 4, wherein the first sensor is an ultrasonic wave ultrasonic sensor and the second sensor is an infrared sensor.
The robot following method according to claim 1, wherein the obtaining the minimum value of each group as a predetermined detection distance according to a predetermined rule comprises:

When the distance of the obstacle is detected to be within the first preset range, the detection distance of the second sensor is taken as the detection distance determined by the group;

When the distance of the obstacle is detected to be within the second preset range, the detection distance of the first sensor is taken as the detection distance determined by the group;

When the distance of the obstacle is detected within a third preset range, the detection distance of the first sensor and the detection distance of the second sensor are merged to obtain the determined detection distance of the group.
The robot following method according to claim 6, wherein the combining the detection distance of the first sensor and the detection distance of the second sensor to obtain the determined detection distance of the group, the fusion method is: d= a×x+b×y, where

d is the distance of the obstacle, x is the detection distance of the first sensor, y is the detection distance of the second sensor, and the values of a and b are determined by the values of x and y, when x is in the range of 0.6 to 0.8, a The value is 0.5 to 1 Within the range, the value of a is proportional to the value of x; when y is in the range of 0.4 to 0.6, the value of y varies from 0.5 to 1, and y is negatively proportional to b. Relationship; when x>0.6, y<0.6, the values of a and b are both 0.5.
The robot following method according to claim 1, wherein the determining the position of the obstacle and tracking the obstacle is performed according to the change of the time and distance of the speed of the robot and the detection of the detection distance by the two sensor modules before and after. : Judging the distance and direction of the obstacle according to the detection distance determined twice before and after, judging the movement state of the obstacle, the speed of movement and the size of the obstacle.
The robot following method according to claim 1, wherein the robot following method further comprises:

Pre-storing a number of specified gestures of the tracking object, and one-to-one correspondence between the specified gestures and the distance between the robot and the tracking object;

A number of specified gestures of the tracking object are identified, and the robot is adjusted to a corresponding distance according to the identified specified gesture.
The robot following method according to claim 9, wherein the one-to-one correspondence between the plurality of specified gestures and the distance between the robot and the tracking object comprises: the plurality of specified gestures include waving, waving, and raising a hand.

When the tracking object is beckoning, adjust the robot to advance to a specified distance closer to the object to be followed.

When the tracking object swings, adjust the robot to retreat to a specified distance further from the object to be followed.

When the tracking object raises the hand, the robot stops at its original position.