CN106940208A - Robot target demarcates the system with oneself state monitoring function - Google Patents

Abstract

The invention provides a kind of demarcation of robot target and the system of oneself state monitoring function, it is characterised in that range measurement and image taking module, robotary data acquisition module, data transmission module, data processing and display module including object.As a result of above-mentioned technical scheme, the present invention compared with prior art, has the following advantages that and good effect：Range measurement, IMAQ, data processing and display, performance monitoring are integrated in one by the present invention, inexpensive, quick can realize that goal seeking and surrounding environment are built.The present invention is what the principle based on Fusion was realized, by experiment test, present system working stability, favorable expandability, low cost, interactive interface friendly, system working stability directly perceived, can accurately, quickly perform some goal seeking tasks in specific environment.

Description

A system for robot target calibration and its own state monitoring function

technical field

The invention relates to a target recognition system based on multi-sensor fusion by using a robot.

Background technique

With the development of social information technology, mobile robots with autonomous navigation capabilities are widely used in various industries. In the process of walking and exploring, the robot must be able to recognize objects in the surrounding environment to realize the movement of the robot in the unknown environment, to find the target according to human requirements and to avoid damage to the robot body and equipment.

At present, many methods have been successfully applied to object recognition based on vision, such as laser ranging, ultrasonic, millimeter-wave radar, multi-source fusion and computer vision-based recognition methods. Target recognition methods based on ultrasonic waves are mostly used in the field of robot obstacle avoidance. The classic ultrasonic ranging technology has a wide range of applications and unique advantages, but in the actual application environment, relying on a single ultrasonic ranging sensor to design poor adaptability, can only accurately measure the distance of obstacles, the distance of obstacles Most other information such as location can only be obtained through fuzzy inference, and its real-time and accuracy are difficult to guarantee. In addition, ultrasonic waves have many limitations in terms of distance measurement accuracy, measurement blind area and measurement range. Although the vision-based target recognition method can achieve large-scale and precise data collection, the richness of sampling information also introduces higher system costs; moreover, the expansion of functional modules requires more powerful microelectronics technology and power supply Therefore, the power consumption of the robot increases, the continuous working time becomes lower, and the real-time performance decreases.

In addition, the embedded robot does not have a set of visual operation interface for the working status. The existing robot remote monitoring system often relies on the monitoring platform provided by the robot manufacturer, which is not only not universal, but also often expensive. Small companies in the robotics industry stay away. At present, the working status of embedded robots is completely dependent on the analysis of the actual work situation by experienced robotics engineers, and has not gone deep into the system level. The CPU workload and detailed real-time running status of embedded robots are even more ignorant. This is not good for the long-term, efficient and safe use of robots.

In order to overcome the contradiction between the function and power consumption of traditional mobile robots, so that the robot can work stably in more complex and harsh environments, the current common method is to separate the mobile system of the robot from the sensor and processor system, and use high-capacity batteries Or long wires are powered by the power grid, and a microcomputer is combined with it to realize corresponding functions. This method is suitable for laboratory or field environment, and can carry out precise distance measurement and complex image processing. But the volume of the corresponding robot is too large, the cost is high, and the degree of automation and intelligence is low.

Contents of the invention

The purpose of the present invention is to provide a robot-based system with target distance measurement, image processing and self-state monitoring functions, which has strong real-time performance, good interaction, low cost, wide application, high precision, and convenient function expansion and good stability.

In order to achieve the above object, the technical solution of the present invention is to provide a system of robot target calibration and self-state monitoring function, which is characterized in that it includes a distance measurement and image capture module of the target object, a robot state data acquisition module, and a data transmission module. , data processing and display module, wherein:

The distance measurement and image capture module collects corresponding data through various sensors and stores the collected data on the robot;

The robot state data acquisition module is used to monitor robot information in real time, and the robot information includes various parameters of the robot and the data collected by each sensor in the distance measurement and image capture module;

The data transmission module transmits the original data obtained by the distance measurement and image capture module to the computer through wireless means;

The data processing and display module, after receiving the data obtained by the computer terminal, processes it into the intuitive target information desired by the user, and displays it on the interactive interface together with the robot information monitored by the robot state data acquisition module.

Preferably, the distance measurement and image capture module includes an ultrasonic sensor, an infrared sensor, and a visual sensor, wherein:

The ultrasonic sensor is used to measure the distance of the target, and the infrared sensor is used to supplement the blind area of the short-range ultrasonic sensor. The ultrasonic sensor and the infrared sensor are evenly distributed on the side of the robot, and the number, distance and general orientation of the target are estimated by the difference in the measured value. ;

The visual sensor is used to capture the environmental information directly in front, and the visual sensor is located on the front of the robot.

Preferably, the information displayed by the data processing and display module includes robot speed, acceleration, target distance, battery usage, CPU load level, and at the same time, it also displays the shape and orientation of the target obtained through image processing, through the visual sensor The size of the target obtained by calibration, and the specific information of the target and its surrounding environment information are obtained by fusing data from multiple sensors.

Preferably, various parameters of the robot include speed, position, distance from the target, battery usage, and CPU load.

Preferably, the data transmission module includes a Wi-Fi communication module, and a network connection protocol and data transmission are established through the Wi-Fi communication module.

Due to the adoption of the above-mentioned technical solution, the present invention has the following advantages and positive effects compared with the prior art: the present invention integrates distance measurement, image acquisition, data processing and display, and performance monitoring, and can be low-cost, fast Enables object finding and surrounding environment building. The present invention is realized based on the principle of multi-sensor data fusion. Through experimental tests, the system of the present invention has stable operation, good scalability, low cost, intuitive and friendly interactive interface, stable system operation, and can accurately and quickly execute some goals in a specific environment. Find tasks.

Description of drawings

Fig. 1 is a block diagram of the present invention;

Fig. 2 is a schematic structural diagram of the distance measurement and image capture module in the present invention;

Fig. 3 is the interactive schematic diagram of robot, local window machine and Linux virtual machine;

Fig. 4 is a schematic diagram of obtaining the distance between the surrounding unknown objects and the robot by using the robot ultrasonic sensor and the infrared sensor together;

Figure 5 is the robot motion model.

detailed description

Below in conjunction with specific embodiment, further illustrate the present invention. It should be understood that these examples are only used to illustrate the present invention and are not intended to limit the scope of the present invention. In addition, it should be understood that after reading the teachings of the present invention, those skilled in the art can make various changes or modifications to the present invention, and these equivalent forms also fall within the scope defined by the appended claims of the present application.

The embodiment of the present invention relates to a set of robot target finding and state monitoring system. The system includes a target object measurement and robot state data acquisition module, a robot state data acquisition module, a data transmission module, and a data processing and display module; the target object measurement The robot state data acquisition module is used to obtain raw data, such as target distance and shape, and record robot parameters to obtain the robot’s running state and estimate the robot’s motion state; the data transmission module uses TCP/IP protocol or The SSH protocol sends the data of each sensor to the computer terminal; the data processing and display module calculates the trajectory of the robot through the real-time obtained speed and gyroscope information, and uses mathematical tools such as matlab to process the obtained original pictures to obtain the robot The shape, color and other properties of the object are captured during the movement, combined with the calibration of the visual sensor to calculate the size of the object, and all the information is counted to obtain the target and the approximate environmental information during the search process.

Fig. 1 is a structural block diagram of a robot object recognition system based on multi-sensor fusion in the present invention. It can be seen from Figure 1 that the system includes the distance measurement and image capture module of the target object, the robot state data acquisition module, the data transmission module, and the data processing and display module.

As shown in Figure 1, multiple embedded robots move in formation, jointly detect unknown objects in different directions, and store the obtained information; install zabbix monitoring software in the robot, local window machine and Linux virtual machine to monitor the robot Its own motion information, and send the data obtained by the sensor to the window machine, use the local machine to process the data, and then send the result to the Linux virtual machine and display it on the front end.

As shown in Figure 2, the distance measurement and image capture module mainly uses the ultrasonic sensor to measure the distance between the robot and the unknown object. Due to the "blind area" of the ultrasonic sensor, the infrared sensor is used to measure the proximity of the robot as compensation; the visual sensor is used to intermittently shoot Image of the robot moving straight ahead. The distance, image and other data obtained by the sensor are stored in the memory of the robot.

As shown in Figure 3, the zabbix monitoring software is installed in the robot, the local window machine and the Linux virtual machine. Based on the wifi mode of the local router, the secureCRT is used to log in to the robot operating system through the SSH protocol, and the robot runs the code and completes it on the robot memory. Data transmission; set monitoring items in the robot, local Windows machine and Linux virtual machine, store the information obtained by the robot sensor in the database of the Linux virtual machine, and display it on the front end.

As shown in Figure 4, the distance between the surrounding unknown objects and the robot is obtained by using the robot's ultrasonic sensor and infrared sensor to work together. The robot adopts CMOS image acquisition sensor, which has the advantages of high sensitivity, low noise, wide spectral response range, fast output image and good dynamic performance, and can capture relatively clear images under the irregular motion state of the robot. Image processing is mainly divided into four parts: image grayscale and binarization, filtering, edge detection and color matching. First, the image is processed in grayscale, and the single-scale Retinex algorithm is used to eliminate the shadows formed by the background and obstacles due to natural light, to reduce the contrast between the shadowed road surface and the non-shaded road surface, and to enhance the image; the scan line seed method is used to fill the area, using Canny edge detection extracts the edge, that is, the outline of the object, and records the information; finally, the color feature matching method is used for the position of the image object to find whether it matches the target color.

The motion model of the robot is shown in Figure 5. The motion of the robot only considers the movement of the two-dimensional plane, and does not consider the slope movement of the inclination angle, that is, the movement in the Z-axis direction of the unmanned vehicle is not considered. Therefore, the pose information q of the unmanned vehicle includes the two-dimensional plane location and orientation.

q=[xy] ^T

In the formula, x and y are the position of the robot in the two-dimensional coordinate system, and are the angle between the direction of the robot and the positive direction of the x-axis. If the initial orientation of the robot is the positive direction of the x-axis.

The linear velocity of the robot is v, then the linear velocity of the unmanned vehicle in the axis direction of the two-dimensional coordinate system linear velocity in and axis direction for:

The gyroscope in the robot can measure three data X, Y, Z: respectively represent the angular acceleration of the robot in the YZ, XZ, and XY planes, and the angular velocity ω can be obtained by integrating the angular acceleration, and the angle θ can be obtained by integrating again. Finally, use the data obtained above to deduce the trajectory of the robot, and combine the object information obtained by image processing to mark the relative position of the object and the robot.

The data acquisition and processing of this example also need to be realized through the corresponding program code, that is, first write the robot motion control related code, so that it can move smoothly and at a constant speed, avoid obstacles and reduce the impact of obstacles on the robot's motion state ; Then install zabbix monitoring software in the robot embedded system, local window system and Linux virtual machine, set monitoring items, write monitoring and file transfer scripts, and display data on the front end; then, use matlab scripts in the local system to process the original image , and put the result into the monitoring object; finally, carry out information fusion on each sensor to obtain the target object.

Claims (5)

1. A system of robot target calibration and self-state monitoring function, characterized in that, comprising distance measurement and image capture module of target object, robot state data acquisition module, data transmission module, data processing and display module, wherein: The distance measurement and image capture module collects corresponding data through various sensors and stores the collected data on the robot; The robot state data acquisition module is used to monitor robot information in real time, and the robot information includes various parameters of the robot and the data collected by each sensor in the distance measurement and image capture module; The data transmission module transmits the original data obtained by the distance measurement and image capture module to the computer through wireless means; The data processing and display module, after receiving the data obtained by the computer terminal, processes it into the intuitive target information desired by the user, and displays it on the interactive interface together with the robot information monitored by the robot state data acquisition module. 2. The system of a kind of robot target calibration and self-state monitoring function as claimed in claim 1, wherein the distance measurement and image capturing modules include ultrasonic sensors, infrared sensors, and visual sensors, wherein: The ultrasonic sensor is used to measure the distance of the target, and the infrared sensor is used to supplement the blind area of the short-range ultrasonic sensor. The ultrasonic sensor and the infrared sensor are evenly distributed on the side of the robot, and the number, distance and general orientation of the target are estimated by the difference in the measured value. ; The visual sensor is used to capture the environmental information directly in front, and the visual sensor is located on the front of the robot. 3. A system of robot target calibration and self-state monitoring functions as claimed in claim 2, wherein the information displayed by the data processing and display module includes robot speed, acceleration, target distance, battery usage, At the same time, it also displays the shape and orientation of the target obtained through image processing, the size of the target obtained through the calibration of the visual sensor, and the specific information of the target and its surrounding environment information obtained by fusing data from multiple sensors. 4. A system of robot target calibration and self-state monitoring functions as claimed in claim 1, wherein the various parameters of the robot include speed, position, distance from the target, battery usage, and CPU load. 5. The system of a kind of robot target calibration and self-state monitoring function as claimed in claim 1, wherein said data transmission module comprises a Wi-Fi communication module, and establishes a network connection protocol and data through the Wi-Fi communication module. transmission.