ROBOTICS

INTRODUCTION               

        Robots are becoming essential and integral intelligent devices that are used to perform a variety of tasks, which sometimes are beyond the scope of human beings. They find extensive use in areas like industrial automation, nuclear installation, and pharmaceutical and medical fields, space research etc. The development of an autonomous mobile robot able to vacuum a room or even an entire firm is not a trivial challenge. In order to tackle such a task, so that it could be completed with some simplifications and also assumptions were made to the designer’s initial idea of an “ideal” autonomous vacuum cleaner. In this way, some functional requirements that would improve the robot performance were not taking into account due to their inherent complexity or to their mechanical implications. Probably the decision that the most affects the robot complexity is the ability of mapping the environment so that it would exhibit a much better efficiency when compared with the minimalist approach as the one followed (random navigation).With the aim of keeping our robot as simple as possible, while able to perform the initial goals, i.e. an autonomous vacuum cleaner robot able to randomly navigate through a room or a house with the minimum human assistance, the following specifications were found:

These specifications correspond to some of the expected behaviors that will be programmed into the robot. Other behaviors that will increase the overall performance of the robot, such as self calibration of the sensors and navigation with some memory (not completely random) were also considered. During robot moving, the obstacle is detected by the two sensors, one is ultrasonic sensor and another one is infrared sensor. Here these sensors give input signal to the controlling unit (i.e., 89C51 microcontroller). According to the input signal coming from these sensors, the controller unit interrupts the drive unit. Drive unit is fitted in the vehicle to drive the mobile platform. The driving unit works according to the signal form the controller unit. Then the vehicle moves left or right based on the obstacle detected by the sensor. Here the vacuum cleaning used by a hand vacuum cleaner, placed over a mobile platform Based on the vehicle movement dust are cleaned.

BLOCK DIAGRAM OF AUTONOMOUS MOBILE ROBOT:

       The block diagram consists of a microcontroller, power supply unit, PC interface unit, LCD display and keyboard control unit, stepper motor driving unit and sensors and signal conditioning unit.

DESCRIPTION OF THE BLOCK DIAGRAM:

          The major units in the block diagram are

1. Sensing Unit

2. Controlling Unit

3. Driving Unit

4. Power Supply Unit

SENSING UNIT:

          The sensing unit has the infrared sensor and ultrasonic sensor. The infrared sensor is employed for path detection. It senses for the presence of obstacle in its path. If any obstacle is detected, the receiver gets a signal and sends it to microcontroller. The robot waits until the obstacle is cleared. Here the sensing range is lesser. The ultrasonic sensor is mounted over a programmable servo turret. Initially the ultrasonic sensor is kept at middle position, when the sensor senses any obstacle, the microcontroller receives the data from the sensor and it will rotate the base plate of the stepper motor. Thus the ultrasonic sensor can sense the dimension of the obstacle. As a result the microcontroller instructs the stepper motor to deviate its path.

SENSORS AND SIGNAL CONDITIONING UNIT

3.3.1. ULTRASONIC SENSORS

            A sonar range finder works by generating a short burst of sound – a “ping” – then listening for the echo of the sound when it bounces off the nearest object.

ULTRASONIC SENSORS

            By accurately measuring the time from the start of the ping until the echo returns back to the sensor, the distance to the nearest object can be determined. Sound travels at 1116.4 feet/second (340.29 meters/second) at sea level. The sound travels to the object and back, so the distance to the object can be calculated by dividing the elapsed time by twice the speed of sound.

OPERATION:

          As shown in above figure (b), a sonar range finder is operated by generating a pulse of greater than 10 microseconds on its trigger input signal. This causes the range finder to issue a ping. The range finder enables its receiver 100 microseconds after the ping and raises the sensor’s echo output signal. The delay in enabling the receiver prevents the receiver from hearing the transmission of the ping. When the receiver hears the echo it drops the output signal. The elapsed time between the ping and the echo can be determined by measuring the pulse duration on the echo line and adding 100 microseconds.

INFRARED SENSORS:

             Infrared sensors consist of infrared transmitter and infrared receiver. The transmitter transmits beam of infrared rays. The receiver works on the principle of phototransistor that when light image falls on the exposed base of transistor, then the collector will allow current through base. Here the detection is done by two methods. In the first method the beam of light rays are passed towards the receiver and the receiver will give continuous current until the obstacle disturbs the light ray.

1.THE OBJECT DEACTIVATES THE RECEIVER

2.THE OBJECT ACTIVATES THE RECEIVER

           In another method, the beam of light rays are passed, if any obstacle disturbs the light rays, then the infrared rays gets reflected towards the receiver and thus the receiver will give voltage output.

IMAGE SENSORS- COPIS:

COPIS SENSOR

         For advanced controlling and path navigation, image sensors are used. Mostly cameras of either color or black and white (low level sensing) are used. Yasushi Yagi, developed Map-based Navigation for a mobile Robot with Omni directional Image sensor COPIS. COPIS sensor is used as omni directional image sensor COPIS (Conic projection image sensor) to guide the navigation of mobile robot. The feature COPIS is passive sensing of the unidirectional image of the environment, in real-time, using a conic mirror. A COPIS is a suitable sensor for visual navigation in a real world environment. They reported a method for navigation a robot by detecting the azimuth of each object in the omni directional image. The azimuth is matched with given environmental map. The robot can precisely estimate its own location and motion because COPIS observes a 360° view around the robot, even when all edges are not extracted correctly from the omni directional images. The robot can avoid colliding against unknown obstacles and estimate locations by detecting azimuth changes, while moving about in the environment. Under the assumption of the known motion of the robot, an environment map of an indoor scene is generated by monitoring azimuth change in the image. Attempts have been made to acquire omni directional informational on the environment, including acoustic sensing, passive and active vision. An acoustic sensor can easily acquire a depth map of the environment around the robot. Passive imaging methods using a rotating camera a five eye lens, a conic mirror or a spherical mirror have been used to obtain omni directional view of the environmental. Although precise azimuth information is available in the omni directional view obtained using rotating camera. A disadvantage of the spherical mirror is that the resolution along radial direction of spherical mirror is poor because the view angle along vertical direction is too wide; structure in an environment with walls and doors in a room appear along the circle boundary of the image, therefore, it is difficult to extract all segments from the image as lengths of edge segments. A safe navigation, the robot must be able to avoid collision and objects and be able  to estimate its own location in an environment. There are two types of navigation: the robots moves in an unknown and known environment .In a known environment, robot can be navigated speedily and effectively using not visual information from the input image but also by makes of the given map of a known environment. Another one method is the robot will move unknown environment, it stops every few meters, images and analyzes the images to find obstacles and its own location.

CONTROLLING UNIT:

         The microcontroller used here is ATMEL 89C51, which is coded in Assembly language. Depending upon the signals from the infrared sensor and ultrasonic sensor, the microcontroller controls the movement of the stepper motor. The stepper motor is connected to the controller through drive circuit.

DRIVING UNIT:

        The stepper motor is of permanent magnet type whose shaft is connected to the wheels of the robot. Here we employed three stepper motors. The motor at the front is used for forward movement and the left right movement. The top motor is used to rotate the sensor. The pulse signals from the controller unit energize the coils of the stepper motor for driving it.

POWER SUPPLY UNIT:

    The power supply unit consists of battery, DC to DC converter, regulator circuit. The power supply unit gives enough supply to the microcontroller, sensing unit, stepper motor

drive and to the vacuum cleaner. The battery low is indicated by alarm and when microcontroller senses it, moves the robot towards charging unit.

TYPES OF VACUUM CLEANERS

CANISTER TYPE

           The Canister type of vacuum cleaner houses the suction motor and filtering system in square or rectangular container. It incorporates a "clean-air" system, known for producing strong suction for use with the attachments. Power nozzle is referred as a "power team." A canister type vacuum cleaner usually contains wheels which allow it to be pulled after the user by its hose.

UPRIGHT TYPE

       The Upright type of Vacuum cleaner is self-contained and has a handle extending vertically from its main case. It contains a revolving brush roll for deep cleaning carpets.

Many upright type vacuum cleaners have on-board attachments for cleaning furniture and hard-to-reach corners. Upright vacuum cleaners are pushed around in front of their users, eliminating the need to pull something behind them.

POWER TEAM TYPE

       The Power team type of vacuum cleaner contains a canister type vacuum cleaner with a motorized power nozzle. The powerful brushing action of the power nozzle, combined with the strong suction of the clean-air suction motor, enables a power team to perform very well deep cleaning carpets as well as cleaning all other types of surfaces.

WORKING OF VACUUM CLEANERS

         The main accessories of vacuum cleaner is the intake port, which may include a variety of cleaning accessories, the exhaust port, the electric motor, the fan, the porous bag and the housing, which contains all the other components.

FORMATION OF VACUUM

          The electric current operates the motor. The motor is attached to the fan, which has angled blades as like propeller blades. As the fan blades turn, they force air forward, towards the exhaust port. When air particles are driven forward, the density of particles and the air pressure increases in front of the fan and decreases behind the fan. The pressure level in the area behind the fan drops below the ambient air pressure level outside the vacuum cleaner. This creates suction, a partial vacuum, inside the vacuum cleaner. The ambient air pushes itself into the vacuum cleaner through the intake port because of the air pressure inside the vacuum cleaner is lower than the pressure outside.

As long as the fan is running and the passageway through the vacuum cleaner remains open, there is a constant stream of air moving through the intake port and out the exhaust port.

PICKING UP THE DIRT

         The suction created by a vacuum cleaner's rotating fan creates a flowing stream of air moving through the intake port and out the exhaust port. The moving air particles rub against any loose dust or debris as they move, and if the debris is light enough and the suction is strong enough, the friction carries the material through the inside of the vacuum cleaner. Some vacuum designs also have rotating brushes at the intake port, which kick dust and dirt loose from the carpet so it can be picked up by the air stream. As the dirt-filled air makes its way to the exhaust port, it passes through the vacuum-cleaner bag. These bags are made of porous woven material typically cloth or paper, which acts as an air filter. The tiny holes in the bag are large enough to let air particles pass by, but too small for most dirt particles to fit through. Thus, when the air current streams into the bag, all the air moves on through the material, but the dirt and debris collect in the bag. The vacuum-cleaner bag can be placed along the path between the intake tube and the exhaust port, as long as the air current flows through it. In upright vacuum cleaners, the bag is typically the last stop on the path: Immediately after it is filtered, the air flows back to the outside. In canister vacuums, the bag may be positioned before the fan, so the air is filtered as soon as it enters the vacuum. The vacuum cleaners pick up dirt by driving a stream of air through an air filter, the bag. The power of the vacuum cleaner's suction depends on a number of factors. Suction will be stronger or weaker depending on:

 

                                               

FACTORS INFLUENCING EFFICIENCY OF ROBOTS:

           

THE POWER OF THE FAN

           To generate strong suction, the motor has to turn at a good speed.

THE BLOCKAGE OF THE AIR PASSAGEWAY

            When a great deal of debris builds up in the vacuum bag, the air faces greater resistance on its way out. Each particle of air moves more slowly because of the increased drag.

THE SIZE OF THE OPENING AT THE END OF THE INTAKE PORT

         Since the speed of the vacuum fan is constant, the amount of air passing through the vacuum cleaner per unit of time is also constant. The size of the intake port is not considered much, the same number of air particles will have to pass into the vacuum cleaner every second. If the port is small, the individual air particles will have to move much more quickly in order for them all to get through in that amount of time. At the point where the air speed increases, pressure decreases, because of Bernoulli's principle. The drop in pressure translates to a greater suction force at the intake port. Because they create a stronger suction force, narrower vacuum attachments can pick up heavier dirt particles than wider attachments. One very popular vacuum-cleaner design is the central vacuum system, turns whole house into a cleaner. A motorized fan in the basement or outside the house creates suction through a series of interconnected pipes in the walls. When the fan motor is turned on and a hose is attached to any of the various pipe outlets throughout the house. The dirt is sucked into the pipes and deposited in a large canister. For heavy-duty cleaning jobs, a lot of people use wet/dry vacuum cleaners, models that can pick up liquids as well as solids. Liquid material would soak paper or cloth filters, so these cleaners need a different sort of collection system. On its way through the cleaner,the air stream passes through a wider area, which is positioned over a bucket. When it reaches this larger area, the air stream slows down, for the same reason that the air speeds up when flowing through a narrow attachment. This drop in speed effectively loosens the air's grip, so the liquid droplets and heavier dirt particles can fall out of the air stream and into the bucket. After you're done vacuuming, you simply dump out whatever has collected in this bucket.

PATH PLANNING AND NAVIGATION

           In the mobile robots the path detection and path planning is much important factor for accurate mobility. The path detection can be done in various methods depending upon the accuracy required and the work space environment.

PROBLEM

       The problem of using these image sensors is that it requires a high power processor of image processing and also requires high memory to store the predetermined path.

PATH DETECTION AND OBSTACLE AVOIDANCE

              The below diagram shows the deviation of pre-programmed path of mobile robot, when it encounters a obstacle in its path. The flow chart describes the working of the robot by sensing the obstacle and moving in programmed path. Here the robot uses three sensors to sense the obstacle. It has two infrared sensors fitted at the left and right corner. The ultrasonic sensor is fitted in the programmable rotary table in the middle. When the robot control is switched on in the autonomous mode, the controller commands the robot to move in a straight path, in the programmed manner

When the robot moves in a programmed path, the sensor senses for the obstacle. As the infrared sensor is for low range sensing, the ultrasonic sensor is used further to locate the size, shape of the obstacle. If any one of the infrared sensors is sensed, then it gives the analog voltage to the ADC where it gives the digital pulse to the controller. Now the controller rotates the base of ultrasonic sensor and the corresponding analog voltage is then converted by ADC and it sends the digital data to the controller. As per the angle and the received data, the controller can locate the size and shape of the obstacle. If the location of obstacle in right is more, the controller orders the robot to turn left and move forward. If the obstacle is cleared from the path, the robot turns right and moves forward. If the robot moved outside the obstacle, then it returns to its path. This procedure can be continued for the presence of obstacle in the left. If the obstacle is in the centre, then the robot can take right path. The loop will be followed

                                

FLOWCHART FOR PATH DETECTION

                                        

CONCLUSION AND FUTURE ENHANCEMENTS:

    This project “Autonomous Mobile Robot” eliminates the drawbacks of various robotics manufacturers. The benefits of autonomous mobile robot couldn’t have an observer, due to the sufficient sensor provided. The sequence of predefined path can be modified. It uses battery, provides an on-board power supply and weightless mobile platform comparing to other robots. It eliminates the human labor for cleaning the cotton in the room. This robot is compact in size and also economical one. The benefit of Autonomous Mobile Robot is the path navigation due to sufficient sensor provided and hence there is no need for camera, personal computer and observer. Actually, the cotton pr the debris is cleaned by the human labor and it may cause more cost. The above problem can overcome by implementing “Autonomous Mobile Robot” in the shop floor. Here in the above designed model, there are some draw backs. The two stepper motors used here creates some speed mismatch and causes deviation in xy path navigation. Also the infrared sensors used here has less sensing range and the rotary platform mounted ultra sonic sensor gives moderate sensing and for large area this deviates lightly from the actual path. The battery used has high consumption and the charging is required frequently. For this above draw backs some more modifications are suggested. In the sensor unit instead of ultrasonic sensor, image sensor of black and white (For low cost and less accuracy) or the color sensor (For high cost and high accuracy). This also requires powerful processors with high memory compatibility for image sensing and image processing. This eliminates the poor accuracy of infra red sensor and the ultrasonic sensor.The low power battery is eliminated by adding a light weight high power battery. Also for this increased weight, stepper motor of required torque is used. Here the problem is that the weight of whole system affects the driving system and the driving system affects the battery used. A trial and error method is implemented to maintain the system stability. Though the system may get complicated, once the work area is configured, and then the whole system can work with more accuracy and eliminates manual work.

REFERENCE:

[1]. Andrew Blake, Gabriel Hamid, and Lionel Tarassenko, “A Design for a Visual Motion Transducer”, IEEE Transactions on Robotics and Automation, Vol. 11, NO.5, October 1995

[2]. Fabrice R. Noreils and Raja G. Chatila, “Plan Execution Monitoring and Control

Architecture for Mobile Robots”, IEEE Transactions on Robotics and Automation, Vol. 11, NO. 2, April 1995

[3]. I.Ulrich, F. Mondada and J.D.Nicoud,” Autonomous Vacuum Cleaner”, Robotics and

autonomous Systems 19, 2007

[4]. Johan Forsberg, Ulf Larsson and Ake Wernersson, “Mobile Robot Navigation using The Range-weighted Hough Transform”, IEEE Robotic and automation magazine, March 1995

[5]. Jurgen Gulder and Vadim I. Utkin “Sliding Mode Control for Gradient Tracking and

Robot Navigation Using Artificial Potential Fields” IEEE Transactions on Robotics And Automation, Vol. 11, NO. 2, April 1995

[6]. Lloyd Greenwald and Joseph Kopena, “Mobile Robot Labs” IEEE Robotics and Automation Magazine, June 2003

[7]. R.D.Schraft, “Mechatronics and Robotics for Service Applications”, IEEE Robotics and Automation Magazine, December 1994