|
随着科学技术的发展，特别是计算机技术的发展，机器人在生活中的作用越来越受到人们的重视。移动机器人技术在工业、农业、航天及空间探测等许多领域都起到了重要的作用，同时也在人们的日常生活中显示了其广泛的应用前景。而智能机器人当中的触觉导航机器人是一类能够通过传感器感知周围环境和自身状态，实现在有障碍物的环境中面向目标的自主运动，从而完成一定作业功能的机器人。近年来，移动机器人技术在工业、农业、航天及空间探测等许多领域都起到了重要的作用，同时又显示了广泛的应用前景。因此，机器人技术已经引起了美、德、意、法、日等多个国家学术界的极大关注，研究工作也开始蓬勃发展了。
Abstract: Along with the development of science and technology, especially the development of computer technology, the role of robot in life more and more get people's attention. In recent years, the mobile robot technology in industry, agriculture, space and space exploration, have played an important role in many fields, and shows a broad application prospect. Therefore, robot technology has caused the United States, Germany, Italy, France, Japan and other countries the academic attention, research work also began to flourish.with the development of science and technology, especially the development of computer technology, the role of robot in life more and more get people's attention. Mobile robot technology in industry, agriculture, space and space exploration, many areas have played an important role, but also in People's Daily life shows its broad prospect of application. And intelligent robot navigation robot is a kind of touch sensors to sense their environment and their own state, and achieve in a environment with obstacles facing the voluntary movement of, so as to accomplish a certain operation function of the robot
1 research purpose and meaning
Along with the development of science and technology, especially the development of computer technology, the role of robot in life more and more get people's attention. In recent years, the mobile robot technology in industry, agriculture, space and space exploration, have played an important role in many fields, and shows a broad application prospect. Therefore, robot technology has caused the United States, Germany, Italy, France, Japan and other countries the academic attention, research work also began to flourish.
In the robot technology, aided navigation system as the eyes of the robot, can let the professional and technical personnel to further understand the internal structure of local parts, to avoid the errors caused by inadequate experience, use of robot has accurate positioning, greatly reduce the work intensity and other advantages can make professional project safer, more reliable and more scientific. So, the design of the excellent aided navigation system is the prerequisite of advanced intelligent robot. In addition, we can also through software programming to implement the quiver, improve the working accuracy and other advanced request, makes aided navigation system has extremely broad application prospects.
And intelligent robot navigation robot is a kind of touch sensors to sense their environment and their own state, and achieve in a environment with obstacles facing the voluntary movement of, so as to accomplish a certain operation function of the robot. Because the navigation function is the third generation of robots, intelligent robots is a very important basic function, and based on the research of the tactile navigation function can provide infrastructure assurance for realizing the intelligent mobile robot and technical support.
2 the development of robotics research
2.1 robot technology development
Started since the 70 s, foreign robot tactile sensing technology research, more than 30 years course mainly divided into the 70 s, 80 s and 90 s three times, the three hours of research and achievements of each have their own characteristics:
Robotics research abroad in the 70 s has become a hot spot, but touch technology research is just produce green shoots. Then the research of the sense of touch is limited to contact with the object or not, the size of contact force. Even if there are some good ideas, developed the sensor is also rarely and shabby.
In the 80 s a robot tactile sensing technology research, development, rapid growth. During this period of sensor design principle and method, made a lot of research, mainly has magnetic, magnetoelectric, resistor, capacitor, thermoelectric, piezoelectric ultrasonic and resistance strain, force, light, the principle and method. Was also started to detect the contact point and area, contact section shape and pressure distribution of tactile array research, developed to detect object shape, size, location, presence, reaction mode and the temperature sensor. But for materials such as surface texture, hardness, viscosity rational qualitative tactile perception has proved much more difficult, these problems also haven't been solved effectively. The touch signal processing mainly focused on image processing technology, adopt the method of computer vision research study from static image tactile array. Due to recognize the importance of dynamic signal, thus developed to detect the stress changes, the sliding and transient contact sensors. Later, in order to follow the research model of active visual foreign researchers began active tactile research. Overall, in the 80 s research can be divided into the sensor, touch data processing, active tactile perception of three parts. Its outstanding characteristic is based on sensor devices research center, mainly for industrial automation. During this period the most main, is also commonly used tactile sensor technology standard is put forward by the Harmon tactile sensor technical requirements, this major is also proposed for the robots used in industrial automation. In the 80 s, many people are very optimistic that with the emergence of commercialized sensors, touch technology is mature, and tactile sensor in the near future will be integrated into the factory of robot system.
In the 90 s continued growth of the research of tactile sensing technology, also began to develop in the direction of multiple. According to the width of taxonomy, the literature of tactile research can be divided into: sensing technology and sensor design, the tactile image processing, the shape identification, active tactile perception, structure and integration. During this period, the most active research is still a new sensor design and construction, such as tactile array development, from the 80 s through high density arrays, this part also includes the sensor data processing and analysis of the experiment. Followed by including integrated sensor 2 drive and control of active tactile perception, it briefly is considered to be the key to the development of tactile sensing technology, including multiple refers to the hand and the design of flexible operation, soft material is also important research aspects. In addition, also includes the application of tactile sensing technology in the field of non-traditional research and used in the research of tactile telepresence teleoperation robot technology.
In the 90 s, tactile sensing technology research and development achievements mainly include:
(1) the study of engineering problems. Such as device packaging, to the deep understanding of the sensor performance and operation, the new material research, the change of soft contact characteristics, etc.
(2) increase the understanding of the effect on the sensor. Such as through the use of several kinds of tactile sensor with different response rate of complementary, make the integrated system can handle a given tasks within a certain range of different contact characteristics.
(3) improve the robot dexterous hand, make more refers to the hand to the operation of the object properties have made considerable progress.
(4) the sense of touch is applied in medical field. Tactile sensor in the field of applied research has concern.
In the 90 s, the general characteristics of touch technology and the sensor is not according to the early people expect penetrated into all fields of industry, but it does not reduce its corresponding research, is only a limited amount of solids in the world in the field of industrial steering of some soft relatively disordered world of natural systems.
Robot sensor is about in the 80 s began to research in China, but due to the limitation of objective conditions, since 1987, when the state 863 plan before slowly pace. Under the 863 plan to support, in the early 90 s tactile sensor research in our country has made great progress. As in the study of tactile sensor array, southeast university implements the piezoresistive type 16 x 16 touch sensitive array and its data processing and tactile image recognition; Beijing institute of technology has been developed with sleep, sleep and tactile sensing system which has the function of the image recognition; Hangzhou institute of electronics industry is developed for multi-sensor gripper close to sleep, compound sensor, sliding contact. In recent years, with the support of 863 project, hefei smart, southeast university, Beijing, hangzhou, Harbin institute of technology, huazhong institute of technology, Beijing institute of information on the tactile sensing technology are made have distinguishing feature each, such as the southeast university studied the three dimensional tactile information retrieval methods; Beijing institute of technology has been developed for tactile telepresence finger touch sensors.#p#分页标题#e#
2.2 in the process of the evolution of robotics with reason
2.2.1 main shortcoming
Tactile sensing technology for over 30 years of research and development while has made big progress, but in this process there are also many obvious shortage, to questions mainly are:
First is the application of tactile sensing technology and lead to obvious lag, most of the research results obtained is just stay in the laboratory stage, compared with visual sensing technology in this deficiency is more obvious. The present situation is about cameras and visual sensor has become the integral parts of a robot system, the related hardware software on the market have a larger supply of goods and configuration using easily, already belongs to the generalization of equipment. Even compared with the force sensor, touch touch technology is also obviously lag behind. In domestic situation, for example, domestic wrist sensor was developed in the early 90 s after to enter the commercial. And at that time, and after a period of time, almost no domestic market on the durable, reliable and versatile tactile sensor.
Followed by touch technology research contents and methods of the defects in the aspects, especially in the 80 s. These weaknesses mainly displays in: the performance requirements of sensor technology is not from the market or from the point of view of engineering application, but determined by sensor research and manufacturer; Industry there is too much from the academic point of view of study the phenomenon of touch, sometimes there are too many imitation of the tendency of human perception mode; In the study of multisensor system of tactile and other sensory complementary value is not enough, as in the study of tactile sensor array type; Despite the sense of touch to identify rules graphics or shape has a lot of research, the recognition of the shape and the actual is automated production preferred the visual way.
The causes of 2.2.2
Causes of the shortage or touch technology is relatively backward mainly have two aspects. First is the foundation of lack, the second is at the beginning of tactile sensor application technology and market positioning. Basic lack of major in mechanism and material, people to understand the mechanism of human touch obviously as sight, hearing. In the way people feel, sight, hearing, smell, have specific corresponding organs, not touch. Have touch function of the surface of the whole body. The touch itself is complex, the difficulty is also the biggest. Robot feeling basically is simulated, so the robot tactile perception on the principle of big difficulties. Meanwhile the existing materials science, manufacturing technology, craft and so on limit, the options available to touch the sensitive materials and its performance is limited, far less than a person's skin, this makes the performance of the tactile sensor is fundamentally limited. Of tactile sensor application location was originally in the 80 s, according to the development of robot technology and of the most promising and has the biggest market in the field of industrial automation applications prediction of the, but in fact this is not a proper positioning. Since the 80 s robot is widely used in manufacturing automation, but tactile sensor is not as people look forward to with robots are widely used in the production automation, investigate its reason mainly lies in the industrial automation environment is structured environment. On the one hand, the tactile sensors don't have a general and reliable, durable and other excellent properties, on the other hand after efforts to study and make the tactile sensor has the identification function of the shape of either in the structured environment don't need to, or less than visual sensor, so tactile sensor without a major market demand. But people's expectations of tactile sensor application in industrial automation still remains in the early 90 s, it shows that this technology and the market positioning is due to improper people due to limited understanding of things by age.
2.3 the trend of the development of robot
If the current international machine in strengthening scientific research, robot generic technology research, and development in the direction of intelligent and diversification. The main research content focused on the following eight aspects:
(1) the optimization of the industrial robot manipulator structure design to explore new high-strength lightweight materials, further improve the load - weight ratio, towards modular and reconfigurable institutions at the same time.
(2) the robot control technology research focus on open and modular control system, man-machine interface more friendly, language, graphical programming interface is under development. The standardization of the robot controller and network and network controller based on PC has become a hot research topic.
(3) the multisensor system to further improve the intelligence of robots and adaptability, the use of multiple sensors is the key to problem solving. Its research hotspot is effective and feasible multi-sensor fusion algorithm, especially in the nonlinear and nonstationary and non-normal distribution under the situation of sensor fusion algorithm.
(4) a single mobile robot and multiple mobile robot system, navigation and positioning is always a difficult problem.
(5) virtual robot based on multi-sensor technology, multimedia and virtual reality and telepresence should technology, realize the virtual remote control of the robot and man-machine interactive.
(6) intelligent robot cooperation system and key technologies in developing to the intelligent robots, multi-robot cooperation system is a kind of technology integration platform that has the coverage, at the same time also is a new field of the robot. Multi-robot system of organization and control mode in largely determine the effectiveness of the system, the main group of multi-agent system structure, mutual communication and consultation mechanism, perception and learning methods, modeling and planning and control aspects of group behavior were studied.
(7) the micro and small robot technology which is a new field of robot research and key development direction, past research in this field is almost blank. So research progress in this field will cause a robot technology revolution, and the social progress and incalculable impact on all aspects of human activity, microminiature robot technology research mainly focus on the system structure, movement mode, control method, sensor technology, communication technology and walking technology and so on.
Imitation and bionic technology (8) : this is the highest realm of robot technology development, currently only do some basic research in some ways. But can foresee, in the 21st century will be a robot in the booming era.
2.4 the goal of robotics research in our country
According to the robot development experience, the present situation at home and abroad, and in recent years, dynamic, and combined with the specific situation of the current domestic economic development, "15" during robot technology should focus on intelligent robots and robotic nanomanipulation machinery and related technology development and applications. Conducted on the basis of the robot the restructuring of the assembly system and its related technology research and strengthen the development of multiple sensor fusion and decision making and control integration technology and application research. Application research goal is to focus on our country has developed for many years the type of teaching and reappearing the early stage of the industrialization of key technologies of the industrial robot, vigorously promote the industrialization process, strive to achieve in "15" at the end of spraying of the industrialization of the robot, welding, assembling, etc.
2.5 the current situation of the robotics research in our country
Industrial robots in our country start from the 1980 s, with the support of the country by ">", "five-year" science and technology research, after 20 years of efforts, has formed some competitive industrial robots, research institutions and enterprises. Has basic mastered the robot manipulator design and manufacture technology, control system design of hardware and software technology, kinematics and trajectory planning technology, production of robot key components, developed a paint spraying, arc welding, spot welding, assembling, handling, such as robot, which has more than 130 sets of spraying robot in more than 20 companies nearly 30 automatic spray painting production line (station) on the scale application of arc welding robot has been applied in automobile manufacturing factory of the welding line. But on the whole, China's industrial robot technology and its engineering application level and foreign than there is a certain distance, such as: reliability is lower than the foreign products; Robot application engineering starts late, narrow application field, technology gap with foreign than production line system; On the scale, our country has been installed about 200 units of domestic industrial robot, about four over ten thousand of the global installed sets. Above reason mainly is not formed the industrial robot, the robot of the current our country production is at the request of the user, a customer, a redesigned, specifications varieties, small batch, parts generalization degree low, delivery cycle is long, the cost is not low, and the quality, reliability, stability. Therefore an urgent need to solve the key technology of the early stage of the industrialization, the products in a comprehensive planning, do well the seriation, generalization, modeling design, actively promote industrialization process.#p#分页标题#e#
In recent years, China's industrial robots and automatic production line of the industrial robot related products production and sales amount has exceeded one billion yuan. The current demand in the domestic market at about 3000 units, annual sales of more than 2 billion yuan. Statistics show that China's industrial robots on the market a total number of nearly ten thousand units, accounting for 0.56% of the global total, including all domestic industrial robots (industry relatively large size of the first three industrial robots) industry concentration 30%, the rest is from Japan, the United States, Sweden, Germany, Italy and so on more than 20 countries. Domestic industrial robot is mainly applied in the domestic market, exports of about 100 units, annual exports of more than 020 million.
Intelligent robot and special robot in China with the support of "863" plan, also made a lot of progress. One of the most prominent is the underwater robot, 6000 meters underwater uncabled results of robot in the world leading level, also developed remote control robot, arms coordination control the robot directly, such as climbing robot, pipeline robot model; Of force sensing in robot vision, sound, touch, sleep on the development and application of basic technologies such as carried out a lot of work, had certain development foundation. But in multiple sensor information fusion control technology, remote control and remote control robot, local autonomy system the development of intelligent and robotic nanomanipulation machinery assembly robot applications is just beginning, with foreign advanced level gap is bigger, need on the basis of the original results, have important to system research, to form the supporting system for practical technology and products, in order to later in "fifteen" standing among the world's advanced ranks. 【 A 】
2.6 the main content of robotics research in our country
2.6.1 type of teaching and reappearing industrialization of industrial robot technology research
(1) joint type, side spray type, top gush type, gantry spraying robot product standardization, generalization, modularization, seriation design.
(2) flexible development: copying spraying robot flexible copying composite mechanism development, copying the servo axis trajectory planning research, control system development, the whole machine safety explosion protection, protective technology development, high speed spray cup spraying technology was studied.
(3) the welding robot (the arc welding and spot welding robots as a series of different load robot, can double as arc welding, spot welding, handling, assembly, cutting operations) the standardization, generalization, modularization and series of product design.
(4) the development of arc welding robot welding seam tracking device with a laser vision: the selection of laser and CCD imaging system, visual image processing technology, visual tracking and robot coordinated control.
5) welding robot off-line programming teaching and workstation system dynamic simulation.
6 electronic industry assembly robot product standardization, generalization, modularization, seriation design.
All landowners mass production robot needed for manufacturing, assembling, testing equipment and tools of research and development.
2.6.2 intelligent robot development research
(1) remote control and local autonomous system structure and control strategy research including modeling, remote-controlled robot model, behavior model, who control the dynamic modeling, graphic simulation modeling, virtual instruments and virtual sensor modeling; The man-machine Shared with man as major body planning and control; Local autonomous control; Multiple sensor fusion technology; Two-way force response control; The establishment of the knowledge base, learning and reasoning method; Advanced control technology of human-computer interaction; Virtual reality (VR) control and control of the relationship between the real world; The structure of the monitoring system.
(2) the intelligent mobile robot navigation and positioning technology research including the system structure of navigation and positioning system; In the structure of environment or unstructured environment navigation and positioning method research; Perception system composition of sensor and information processing system; According to the sensor data modeling method of the environment; Fuzzy logic reasoning methods used in research of mobile robot navigation.
(3) the remote control robot system for virtual reality including the man-machine interaction pattern formation and its program design; (carrier and remote control robot manipulator) geometric dynamic graphical modeling; Remote control environment graphical modeling; A remote-controlled robot operation and data acquisition; Virtual sensors and two-way force response, feedback control based on virtual sensors; Task oriented virtual tools; The theory and method of remote control based on virtual reality. Based on VR model and the real world operation can switch, compatibility and interchangeability; VR monitoring system.
(4) the man-machine interaction environment modeling system including CAD modeling of human-computer interaction technology; The interaction in the process of technology of knowledge model artifacts; The layout of the robot and the environment and function verification of interactive technology; Interaction in the sensor data processing technology; Robot calibration modeling, kinematics, dynamics modeling of interactive technologies.
(5) multiple robot remote control technology based on computer screen, including 3 d visual modeling; Model of computer display; Remote control robot model control; Man-machine interface; Network communication.
2.6.3 robotic nanomanipulation machinery research and development
(1) parallel mechanism of machine tool (VMT) and robotic nanomanipulation processing center (RMC) development research including VMT and implementation technology RMC intelligent structure; VMT and RMC key transmission technology; VMT and RMC processing, assembling, placement, coating, testing operation technology; VMT and RMC monitoring detection technology development; With MRC VMT intelligent open CMC control system development; The system software and application software development; Intelligence agencies, materials, mechanical and electrical integration technology; Job status variable intelligent sensing technology; Mechanical and electrical integration of multi-function and smart terminal; General intelligent open CNC control hard software system; Parallel mechanism kinematics and dynamics theory; RMC intelligent control theory; VMT and RMC typical application engineering development.
(2) robotic nanomanipulation unattended and more machine remote control with adaptive ability of large bulk material conveying equipment including monitoring bulk conveying system and sensor fusion and configuration of remote control technology; Fieldbus intelligent sensor technology; Robot motion planning in the application of amount of material piling and taking, autonomous operation; Remote real-time communication based on wide area network; With functions of monitoring and management system of fault diagnosis.
2.6.4 based on robot assembly system restructuring
1) open modular elements including general assembly robot extraction; Special pieces of standardization; Module assembly robot CAD design; General mainstream computer controller structure; Man-machine interface; Network function.
(2) the robot assembly oriented design techniques include digital assembly and CAD integrated technology; Product robotic nanomanipulation assembly planning generation technology; The fuzzy evaluation for assembling products.
(3) the robot flexible assembly system design technology including unit: feeder system intelligent design, the end of the actuator rapid implementation, logistics transport and its control and communication; Integration technology, the flexible assembly line simulation software, the management system.
(4) reconfigurable robot flexible assembly system design technology based on task and environmental dynamic reconfiguration of the robot flexible assembly system theory research; Distributed control system based on the commune (Agent) technology and system of collaborative planning between each element.
(5) assembly force sensing, visual technology, including high precision, high integration of six dimensional wrist sensor technology; Visual identification and positioning technology.
6. Intelligent assembly strategy and its control including assembly state real-time detection and monitoring; Intelligent assembly sequence and path planning and control technology.
2.6.5 multi-sensor information fusion technology and configuration
(1) the robot's sensor configuration and fusion technology in the cement production process control and application in sewage treatment automatic control system including process oriented and configuration of multi-sensor fusion technology; Fieldbus intelligent sensor technology; New sensors developed for technical requirements.#p#分页标题#e#
(2) mechanical and electrical integration of intelligent sensor including perception, voluntary movement, the clean-up (self-tuning, adaptive) mechanical and electrical integration of the sensor research; Process requirement oriented motion mechanism design, implementation, testing and clean-up of voluntary movement; The adjusting control system; Robot mechanism and control technology in the application of the sensor design.
3 The basic content of research and solution
My topic is "touch navigation robot research and design", the overall scheme is to design a control robot, in a simulated flat structure of movement in the rooms, mobile robot by tactile perception noumenon, in order to achieve autonomous mobile, obstacle avoidance, traverse the room navigation functions such as the required basic functions. On the problem of obstacle avoidance, the main focus is to study the motion of the robot obstacle avoidance algorithm, to design a simple, accurate and fast algorithm of obstacle avoidance process, and this function is affected by multiple factors, including the foundation and simulation of furniture. In the navigation at the same time, however, have to implement touch navigation robot as short as possible time to complete a task, is a kind of hard to avoid problems. Because of the way based on the sense of touch to navigate and means when entering a "trap", may have is there is no room for back. Superior only obstacle avoidance algorithm does not solve robots under special circumstances of "self", and thus lead to robots in a scene repeated actions to be from the "trap" inside out. So you need to use a variety of means or different ways to ensure that the robots into a "trap" to "save". Such as household vacuum cleaner robot in the process of cleaning the room, is to use the robot's own more sensing system, motion control system and multiple system functions such as autonomic escapes system.
Mobile robot is a complicated work environment with the planning, self-organizing, adaptive ability of robot. , in the study of the relevant technology of the mobile robot navigation technology is the core technology, as well as the real intelligent and completely, the key technology of autonomous mobile in the navigation path planning has played a very important role. In the behavior control of robot, there are a variety of different control mode, here to name just two more mature control strategies.
"Motion control strategy based on fixed model" thought is the central control is given priority to, the behavior of the concrete is established beforehand library, then through the feedback from the sensor to change all kinds of specific behavior to achieve some degree of environment adaptability. Some this kind of strategy also has revised according to the feedback of movement of original model, such as robot control strategy based on the genetic algorithm. Such strategy has the advantage of design is relatively simple, in terms of control model this kind of strategy is trying to create a single FSM (finite state machine, the finite state machine) to simulate biological highly complex, and the principle has not been fully proved the movement of the decision-making process, which determines the actual performance of this kind of model. Due to the limited state and preset specification is limited, can only on this model in a known environment action, can only have limited adaptive ability.
"Motion control strategy based on behavior" generally does not include the main control computer, system movement completely by each of the joint distribution of the local controller coupled to complete each other. This is a kind of according to the theory of behavioral control design of the control system. Lower organisms are often rely on the interaction of each ganglion to complete the movement, rather than the brain central control (in fact, many creatures don't have the brain or brain is not very developed) also can achieve very high adaptability. This can be considered a without thinking about intelligence: based on the "without thinking of intelligent control system has the characteristics of real-time and self-organization, in a highly uncertain situation has good adaptability and unstructured environment. This kind of control mode the disadvantage is that the state machine complex coupling network design, and is not suitable for some highly uncertain task, and poor repeatability.
4 The hardware system is designed
4.1 the selection of single chip microcomputer
A can work and realize the basic function of computer including CPU, RAM, ROM and the input/output devices, and in the single chip microcomputer, the parts are integrated in a integrated circuit chip, hence the name single chip. Single chip microcomputer to normal work, also has the stable power supply, crystals, external storage and programming debugging interface. The single chip microcomputer work normally required components are encapsulated in a module, microcontroller, micro controller can be directly connected to the PC program development.
Although compared with the personal computer, notebook computer, the function of the single chip microcomputer is very simple, but in the daily life and production activities of ordinary, does not necessarily need to be very complex very advanced computer to implement some functions, such as air conditioning refrigerator temperature control. So, for our customers and users of these products, whether to meet our requirements, whether has a good cost performance is the most important thing. Because of SCM has small volume, light quality, price is low wait for an advantage, make SCM slowly permeated all areas: our bus IC card, civilian luxury car security systems, video camera, automatic washing machine, electronic pets, etc., not to mention all kinds of robots in the field of automatic control, intelligent instruments and medical equipment.
In the single chip microcomputer, 8051 is the most typical of the product, it belongs to the Intel company produces a series of single chip microcomputer is a variety of MCS51. Many companies after the access to Intel technology licensing, on the basis of 8051 as the core of single chip microcomputer of increase or decrease on the function, creating a more typical of a single-chip computer AT89C51 (C51).
This design USES the single chip AT89S52 is the model, it is modified based on AT89C51. AT89S52 is a high-performance, low power consumption of 8-bit MCU, it contains 8 k bytes ISP can wipe again and again 1000 times of Flash memory read-only applications, device adopts high density of ATMEL company, nonvolatile memory technology, compatible with standard MCS51 instruction system and its pin structure. In practical engineering applications, the powerful AT89S52 devices has become many cost-effective solution for embedded control system application. AT89S52 devices do not require simulator and programming machine, only need to use your ISP download line to MCU Flash repeatedly wipe, so easy to use, and flexible configuration, scalable.
4.2 the robot components connection mode
2 components
For the convenience of single chip microcomputer and power supply, the ISP download cable, serial port, and a variety of sensors and motor connection, so you need to make a circuit board, and put in the MCU circuit boards, this is the teaching board, as shown in figure 4-1. The design of navigation robot, namely USES the AT89S52 single chip microcomputer as the brain, through teaching board installed in the robot's turf, after the completion of the robot images as shown in figure 4-2.
Figure 4-1 teaching board
Figure 4-2 robot using C51
C51 teaching board needs to connect the power supply to run, but also need to connect to the PC to complete the programming and interactive work. As the attachment to complete later, you can start using the editor software to test the robot system.
4.2.2 a serial port connection
Robot teaching board to the PC via a serial port cable connection interact with the user. If a PC with serial interface can use a serial port to connect the cable directly, without the need to use the USB serial port adapter, as shown in figure 4-3.
Figure 4-3 USB serial port adapter
Holdings of ISP download line connection
Robots through the connection to the ISP download line in the PC parallel port (as shown in figure 4-4) to download the teaching board on the single chip microcomputer. The ISP download line big end connected to the PC's parallel interface, small end connected to the teaching program download mouth.
Figure 4-4 ISP download line
4.2.4 installation and electricity of the battery
This design made the robot USES is alkaline batteries to power a robot clicks and teaching board, ensure that the polarity of the batteries in the correct direction load. In teaching floor there is a three position switch, when the switch is located in the "0" position, then the device is in the closed state. Will now be three switch by "0" to "1", open teaching board power supply, as shown in figure 4-5 a. Check whether the teaching floor green LED power light is brighter. If not, check the battery in a battery pack and battery box joint is inserted into the teaching board outlet. After the switch to "2", as shown in figure 4 to 5 b, when the power is not only to teaching board power supply, at the same time also will give the robot's executive arm, servo motor power supply, the same, at this time of the green LED power light will brighten.#p#分页标题#e#
Figure 4-5 a closed figure 4-5 b of the three switches in the three one of the state of the switch
4.3 the robot assembly steps
4.3.1 robot chassis hardware installation
1) ready to chassis hardware, confirm the number of parts: 1 baby car chassis, stud 4, pan head screw 4, 13/32 of an inch of rubber ring 1, as shown in figure 4 to 6.
(2) plug the rubber ring of 13/32 of an inch to the hole, the center of the baby car chassis at the same time ensure that the edge of the chassis center hole embedded in the rubber ring grooves;
(3) the stud with four screws as shown on the chassis.
Figure 4-6 chassis parts and part of the installation diagram
4.3.2 dismantle servo throat
Disconnect the power of the controller and the servo motor,
(2) remove all the batteries in the battery box;
(3) the servo motor from your teaching floor disconnected;
(4) use a screwdriver to remove the connection between servo throat and motor output shaft screw;
(5) each quad parts removed from the motor output shaft, as shown in figure 4 to 7;
6. To save screw is good, also can use later.
Figure 4-7 quadrangle parts removed schematic diagram
4.3.3 the motor installed on the chassis
1) ready to install use parts: robot chassis (partially assembled), a continuous rotary motor 2, 3/8 of an inch screw eight, eight nut;
(2) the motor is fixed with screw and nut on the chassis, notice shall not put the machine into the chassis from the outside from the inside of the rectangle window, in order to ensure the best performance.
(3) with the label paper identification of servo motor wheel around, as shown in figure 4-8.
Figure 4 to 8 installed motor parts and schematic diagram
4.3.4 install a battery pack
(1) ready to install the material: the robot chassis has been partially assembled), 3/8 of an inch, 4-4-40 flat head screw, nut, 40 with a plug of the battery box;
(2) with a flat head screw and nut the battery box fixed below the baby car chassis of the robot, as shown in figure 4-9 on the left, the screw through the battery box, and then on the chassis with nut fastening;
(3) the servo motor and battery pack power line line through the chassis with rubber ring hole among them, as shown in figure 4-9 on the right.
Figure 4-9 battery box installation schematic diagram
4.3.5 installed wheels
: (1) ready to install material part has been shipped a good robot, 1/16 "split pin wheels, tail wheel, rubber, plastic ball, save in the demolition of servo throat steps screw;
(2) installed on the chassis of the tail wheel is a central hole of the plastic ball, with split pin shaft, wheel fixed on the chassis, and make sure that the wheels of the center hole and the center hole at the end of the chassis aligned in a horizontal line, then split pin holes through the three at the same time (right) to the left of the chassis and tail wheel, chassis, add one end of the split pin bend it won't slip out from the hole, as shown in figure 4 to 10.
Figure 4-10 tail wheel
(3) stretched rubber band, put it on each wheel set, don't pay attention to strength too much so as not to damage the rubber band;
(4) to press the two wheels grooves on the output shaft, ensure that two axes highly consistent with normal occlusion, the groove of the wheel with screw will finally wheel fixed on the output shaft;
4.3.6 installation teaching board to the chassis
(1) ready to install the material: the robot chassis has been partially assembled), 1/4 "4-40 flat head screw, teaching with controller board;
(2) of the servo motor control line is connected to the teaching on the two control interface board, with "L" label will plug connected to the PC3 port, with the "R" label attached to PC2 port plug, which is to ensure that control the output of P1_0 pin is used to control the right of the servo motor and control the output of P1_1 pin is used to control on the left side of the servo motor, connection schematic diagram as shown in figure 4 to 11.
Figure 4. - 11 servo motor and the principle diagram of the circuit board floor attachment
(3) put the teaching board on the chassis four stents, make its aligned with the four holes on the chassis, ensure teaching board close to the drive wheels at one end rather than the end tail wheel, with a flat head screw connection board and support;
(4) to the rest of the line through the rubber ring servo motor and battery line out from under the chassis, roll up the rest of the line between servo motor and chassis, all kinds of wire and put into place, as shown in figure 4 to 12.
Figure 4-12 installed robots.
4.3.7 installation beard
1) ready to install the material: 2 roots of wire, two flat M4-40 screws, 2 1/2 level "circular casing, two nylon washer, 3 - pin - male connector, two 220 Ω resistance (red - red - brown), two 10 k Ω resistance (brown - black and orange), as shown in figure 4-13;
Figure 4-13 beard part drawing
(2) the two screws (M3x22) through the M3 nylon washer and 13 mm circular column, through the motherboard of cylinder, screw into the board under the support of teaching, but do not be too tight;
(3) to one of the beard of wire hook on the nylon washer, another hook under the nylon washer, adjust the position they cross cross but not contact, as shown in figure 4-14 on the left. Building beard circuit diagram as shown in figure 4-14 on the right.
Figure 4-14 beard installation schematic diagram and circuit principle diagram
5 the software design
5.1 overall plan diagram and analysis system
In the process of the robot forward, if encounter obstacles right, the robot will successively, 90 ° left again; If encounter obstacles left, robot will successively, 90 ° turn right; If both root beard run into obstacles, the robot will successively, back again left turn 180 °, and then move on. The above is the system overall scheme diagram (as shown in figure 5-1) words.
Figure 5-1 system diagram of the overall scheme
5.2 the robot servo motor control signals
5.1.1 control signal sequence diagram
As shown in figure 5-2, high level for 1.5 ms, low level for 20 ms, and then repeatedly control pulse sequence, the pulse sequence is sent after the zero calibration of servo motor and servo motor does not rotate. (if the motor rotation, calibration)
By on, the control of motor speed is the duration of the high level, when the high level of 1.7 ms, motor at full speed rotating counterclockwise, as shown in figure 5-3. When the high level for 1.3 ms, motor clockwise rotation at full speed, as shown in figure 5-4.
Figure 5-2 motor speed control signal with zero sequence diagram
Figure 5-3 1.7 ms continuous pulse sequence to make motor at full speed rotating anticlockwise
Figure 5-4 1.3 ms continuous pulse sequence to make motor clockwise rotation at full speed
5.1.2 beard sensor principle
By the beard circuit schematic diagram (i.e., as shown in figure 4-11 on the right), on the right side of the beard state input is done through the fourth foot P1 mouth, left the beard state input is done through P2 port third foot. And every beard is a mechanical often earthing switch (beard grounded because plate peripheral gold-plated hole are connected to the GND). To connect to the beard I/O pin voltage is 5 v, when the beard is touched, I/O short received, so I/O pin voltage to 0 v.
5.1.3 through programming the robot navigation based on the beard
Robots walk forward in the process, encounter obstacles, robots use it one or two root beard to detect obstacles, once the beard to detect obstacles, son call the navigation function makes the robot back or rotate, and then to walk forward, until meet another obstacle.
The following code is to achieve this function:
# include < Boebot. H >
# include < uart. H >
Int P1_4state (void)
{
Return (P1 & 0 x10)? 1-0.
#p#分页标题#e#
}
Int P2_3state (void)
{
Return (P2 & 0 x08)? 1-0.
}
Void Forward (void)
{
P1_1 = 1;
Delay_nus (1700);
P1_1 = 0;
P1_0 = 1;
Delay_nus (1300);
P1_0 = 0;
Delay_nms (20);
}
The void Left _Turn (void)
{
Int I
For (I = 1; I < = 26, i++)
{
P1_1 = 1;
Delay_nus (1300);
P1_1 = 0;
P1_0 = 1;
Delay_nus (1300);
P1_0 = 0;
Delay_nms (20);
}
}
Void Right _Turn (void)
{
Int I
For (I = 1; I < = 26, i++)
{
P1_1 = 1;
Delay_nus (1700);
P1_1 = 0;
P1_0 = 1;
Delay_nus (1700);
P1_0 = 0;
Delay_nms (20);
}
}
Void Backward (void)
{
Int I
For (I = 1; I < = 26, i++)
{
P1_1 = 1;
Delay_nus (1300);
P1_1 = 0;
P1_0 = 1;
Delay_nus (1700);
P1_0 = 0;
Delay_nms (20);
}
}
Int main (void)
{
Uart_init ();
Printf (" the Program Running! \ n ");
While (1)
{
If ((P1_4state () = = 0) && (P2_3state () = = 0)) / / two whiskers touched at the same time
{
Backward (); / / backward
Left_turn (); / / to the left
Left_turn (); / / to the left
}
Else if (P1_4state () = = 0) / / right beard touched
{
Backward (); / / backward
Left_turn (); / / to the left
}
Else if (P2_3state () = = 0) / / left whiskers touched
{
Backward (); / / backward
Right_turn (); / / to the right
}
The else / / beard not met
Forward (); / / forward
}
}
5.3 decision making artificial intelligence
Maybe we will find that such a situation: when the robot into a special corner of the room, he was left whiskers touched, so it turn right back and 90 °, after marching forward right tentacles have been touched, so it back and left 90 ° and keep going after left whiskers was touched, and the robot is so so, and unable to escape into a "trap". But this situation is not we caused by programming errors, the cause of this result merely is that our code is not very perfect.
The key to solve this problem is to record the total number of alternating pressure beard, the most important thing is that must make the application remember every beard in what is the last time under pressure in a state, and makes comparison and the current state of the press. If the state of the before and after, on the other hand, then add 1 in total. In addition, if the total number of more than the prescribed in the program of the threshold, you will need to make robot do a U turn, need to reset the beard alternating counter at the same time.
Now to specific analysis of the "corner trap" : if the obstacles ahead is a corner of the room, the robot's two tentacle will touch the left wall and right wall alternately, when tentacles touching the left wall, P2_3state () function returns 0, P1_4state () function returns 1. When the tentacles touching the right wall, P2_3state () function returns 1, P1_4state () function returns 0. Which can make the following conjecture: if the obstacles ahead is a corner of the room, then this function returns the value of the two will not equal to each other, and two times before and after each function return values are not equal, then we can set two variables in the program is used to record two tentacles first touches the old state, and then compared with the current state of the two tentacles, if the state of the first and the second appear alternately, alternating with a new variable to record the counter the number of times, when the total number of alternate exceeds the preset threshold in the program, you can let the obstacles in front of the robot "that" is the corner of the room, so as to make corresponding "intelligent" actions. Through programming the car make a u-turn, turned to flee the corner, at the same time reset alternating frequency counter variable, the following is specific to this line of thinking to write the main function, make the robot can escape from the corner, the main function is as follows:
Void main (void)
{
Uchar counter = 1; / / total number of tentacles collision
Uchar old_R = 1; / / right tentacles old state
Uchar old_L = 0; / / left whiskers old state
While (1)
{
If (P1_4state ()! = P2_3state ())
{
If ((old_R! = P1_4state () && (old_L! = P2_3state ()))
{
Counter = counter + 1;
Old_R = P1_4state ();
Old_L = P2_3state ();
> 4) if (counter
{
Counter = 1;
Backward (); / / backward
Left_Turn (); / / to the left
Left_Turn (); / / to the left
}
}
The else
Counter = 1; / / not getting trapped in the "trap" corner
}
If ((P1_4state () = = 0) && (P2_3state () = = 0)) / / two tentacles encountered at the same time
{
Backward (); / / backward
Left_Turn (); / / to the left
Left_Turn (); / / to the left
}
Else if (P1_4state () = = 0)/met/right tentacles
{
Backward (); / / backward
Left_Turn (); / / to the left
}
Else if (P2_3state () = = 0) / / left whiskers touched
{
Backward (); / / backward
Right_Turn (); / / to the right
}
The else / / tentacles didn't meet
Forward (); / / forward
}
}
Doing this program analysis: a careful look at the main function, it is not hard to find, first new defines the three variables in the program
Uchar counter = 1; / / total number of tentacles collision
Uchar old_R = 1; / / right tentacles old state
Uchar old_L = 0; / / left whiskers old state
The variable, the number of the counter is used to record the counter alternating initial value is set to 1; Old_L, old_R respectively used to record two tentacles around the old state values, old_R initial value is set to 1, old_L initial value is set to 0, when tentacles in the corner touch alternately value accumulated to 4, the counter is reset to 1.#p#分页标题#e#
According to the mentioned above, the first step is to check: is there a and only a whisker was touched. The easiest way is to ask P1_4state () whether the return value and P2_3state () return values are equal, the specific statement:
If (P1_4state ()! = P2_3state ())
After the tentacles are touched, the next thing to do is to check whether the current state of the antenna is different from last time. That is to say: when old_R is not equal to P1_4state () and old_L is not equal to P2_3state (), the tentacles touches on the counter plus 1, the current state of the note at the same time, the current P1_4state () returns a value assigned to old_R, current P2_3state () returns a value assigned to old_L, namely:
If ((old_R! = P1_4state () && (old_L! = P2_3state ()))
Counter = counter + 1;
Old_R = P1_4state ();
Old_L = P2_3state ();
> 4) if (counter
{
Counter = 1;
Backward (); / / backward
Left_Turn (); / / to the left
Left_Turn (); / / to the left
}
}
If alternate > 4 counter counter, counter is reset to 1, and perform Backward () function, perform Left_Turn twice () function, control the car to do a u-turn, escape "corner trap".
As for the back of the the else statement is used to deal with wheeled robot education without into "trap" corner, so it is need to alternate reset counter value is 1. To the back of the program is just a general sense of touch obstacle avoidance.
6 Research
Through the study and research of C51 to c programming language design a qualified navigation robot tactile, allows it to be in a structured environment (labyrinth) by means of tactile perception to achieve the function of autonomous navigation. At the same time, the robot in certain scenarios, such as u-shaped channel, T, etc., can be timely correct running of the program, that can in the shortest possible time to leave "trap".
Main reference
[1] www.docin.com/p-151065361.html
[2] the navigation of mobile robot control system based on ARM research [D]. Save the ground: dalian university of technology, in December 2007
[3] http://wiki.dzsc.com/info/2723.html
[4] http://www.kongzhi.net/news/ndetail.php? P = 35494
[5] http://www.iianews.com/ca/_01-ABC00000000000146959.shtml
[6] http://www.chuandong.com/cdbbs/2008-4/28/084280EE759E01.html 【 A 】 zhi-xi Chen. Design and research of the control system of robot arm and [D]. Save the ground: Beijing university of posts and telecommunications, 20050325
[7] pei-pei xu. Complex global path planning algorithm for mobile robot in dynamic environment research [D]. Save the ground: Beijing university of posts and telecommunications, February 18, 2009
[8] http://www.robotdiy.net/index.php? Option = com_content & view = article&id = 109
[9] Xu Quan Wu kratos Si-min li. The infrared wireless transmitter of PPM type based on AT89S52 single chip design [J]. Journal of foreign electronic components, 2006, 5:4 to 8
[10] http://wenku.baidu.com/view/0c794a49e45c3b3567ec8b1f.html
|
 |
|||
| 网站地图 |
