Vision Based Mobile Robot Navigation System

The purpose of the vision system is to recognize the circle as landmark and identify the distance and orientation in a series of reference target image automatically from the closed-range navigation. In order to identify position in an indoor environment, a mobile robot requires attention to characteristics of a target which is to allow mobile robots to make inference in which the distance and orientation of the target and a robot. In this study, fuzzy logic is applied to generate target trajectory movement with the information extracted from vision system such as the distance and the orientation of target. The fuzzy logic controller (FLC) which computes the required speed and angular speed needed by the two motors to drives the robot for target trajectory.


Introduction
The navigation system plays very an important role and challenging competence for mobile robot. Navigation of mobile robot includes a variety of theories and technologies such odometry technique, ultrasonic mapping, and vision system. In the application of navigation, it consists of two areas: global navigation and local navigation [1]. Global navigation such as GPS (Global Positioning System) and INS (inert ial Navigation System) is often used in various open areas. Meanwhile, local navigation such as vision techniques are very effective base in the closed range navigation. There are many applications in indoor environment vehicles and mobile robots. Navigation using vision approach has many tasks, including target matching, target identificat ion and others [1]. Practically, the use of optical sensors and image p rocessing are the major factors affecting the accuracy of navigation [1]. In navigation application, a mobile robot must interpret its sensors data to extract environ ment informat ion, with which the mobile robot can determine its position [2]. After mobile robot can localize its position, it must decide how to act to ach ieve its purpose. Lastly, the mobile robot must control its drive system to achieve the desired path [2] [3]. The increment distance of the robot"s movement sensed with wheel encoders is integrated to calcu late position. Because measurement errors fro m encoders are integrated, the position error is accumulated over time. Using additional sensors can help to reduce the cumulative erro rs.
There are many erro r sources o f d ifferential encoder

Vision Sensing
This work presents an approach using computer vision which applied feature matching technique after the image has been proceeds by Canny edge detector. Features like circular marking, the distance between two vertical lines used by the fuzzy system to analyze whether it has relation with the door being analyzed to design an expert system to detect rectangular shaped door. In th is study, vision system is applied to ext ract the in formation such as the distance of target and the orientation of target.

Canny Edge Detector
In this study, the value of threshold 0.95 for Canny edge detector was chose for both low threshold and high threshold. This process provides a continuous edge and eliminates edge the streaking significant edge [5][6].

Circle Detection
The circle detection technique used in this study is a features matching technique. This technique has two steps, features extract ion and representation between the features. Feature extraction is a process of extracting high-level geometric features of the curves obtained by curve ext raction process. The curves and segmentation of the curves are used to create circular features.
A curve is a set of edge points that are connected to form a continuous contour. Curves usually represent a boundary of the part in the image. The curve extract ion process consists of three steps of finding seed point curves, tracing the curve, and refining the curves.

Distance and Orientati on of Circle
By using a vision sensor to calculate the target distance (D t ) and circle orientation (θc ) as shown in Equation (1) and (2) that refer in [7]. By knowing the diameter and circle center coordination of the circle, the distance and orientation of circle can be calculated. Therefore, the information distance from the target mobile robot can be known through the image circle diameter.
To align the target in the world coordinate frame is to use the actual object size and the computation of object position. The set-up for this test is shown below in Fig.1. The camera is placed directly in front of the target circle. Essentially, by knowing the diameter o f the circle and the calibrat ion of the camera object distance can be measured by the length the image circle d iameter. The distance between the vision sensor and the landmark is inversely proportional, k to the length the image circle diameter. Diameter circle, x' is used to calculate the distance door-to-mobile robots. The calculation used is present in following equation: The Equation (2) was used for the angle orientation of the landmark. The Circle landmark orientation θ c by vision sensor can be calculated using the x-axis pixel position of the center point is calculated at the circle marking. The center circle coordinate, x center is used to calculate the angle of the circle position in the grabbing image using n and c which are constants derived fro m the camera calibration.

Mobile Robot Control System
The posture provides information about how the mobile robot moves with respect to the floor. The speed relative to the ground of the right wheel and the left wheel can be expressed as follows, respectively.
The speed and the angular speed of a mobile robot are related to the wheel speeds, and it can be exp ressed as follows:

Fuzzy Logic Controller 
The master controller is a fu zzy logic controller (FLC) which co mputes the required speed and angular speed needed by the two motors to drives the robot for target trajectory.
A block d iagram of the fuzzy controller is shown in Fig.3. The desired θ d and target distance D t are acquired by calculating the position of the target in the image representing the environment detected by webcam camera. The command signal θ d and D t are transmitted fro m the vision system to the fuzzy controller inside the PC. The error between the co mmand signal and the actual position, as well as the change in error of signal are calculated and fed into the fuzzy controller embedded in the Data Acquisition (DAQ). Fro m the Equation (1) and Equation (2), it can be seen that the differences between the speed of the right and left wheels determines the turn speed. The fu zzy controller is designed to output pulse width modulation (PWM) signal corresponding to u R and u L to the right motor and left motor respectively to control the mobile robot turn angle θ to the desired angle θd and target distance D t.  The rule base stores the rules relating to the knowledge input output relationship of the offered fuzzy controller. The inference mechanis m is accountable for decision making in the control system using estimated sensors that give the informat ion of the target object [8]. The operation involved in this thesis is "AND" because two inputs D t and θ d are involved. Fig. 6 and Fig. 7 show the output membership functions. Note that the output PWM signals should be selected to meet the convergence requirements. The membership of the input and output was designed based on the training of mobile robot modelling it based differential drive technique.  The control rules are designed based on expert knowledge and testing. For example, if θ d is "poslarge" and is D t "far", then the left motor should be much than the right motor, i.e., u L -u R should be "more left" and speed u L +u R should be "fast". Based on knowledge, 15 ru les were obtained. Table 1 represents how to control the turn angle g iven the angle and distance as inputs. The input and output linguistic variables are shown in the table.

Experiments and Results
Experiment results from d istance and orientation calculation are shown in Table 2 and Table 3. Absolute and relative errors are also presented. The relative error was calculated using as following equation: Although the vision system is only using a single webcam, distance calculation of target showed quite accurately. The average absolute error was 0.093 meter and the average relative error was 3.26% for distance measurement. Given the measurement angle of orientation, it also revealed quite accurately with a mean absolute error of 3.16, and the average relative error of 15.66%. Referring to [7], the comparison can be made that Equations (1) and (2) can be used not only in the target in shape of a rectangular but can also be applied to the target in the shape of a circle to enable the calculation of the distance and orientation of objects using monocular vision.
Due to the design of the landmarks and the usage of single webcam, the preliminary step of calibration is necessary. This calib ration has to be done for the extraction of distance and orientation form an image. The usage of only one image makes it mo re difficult to have extremely accurate results although mathematical geo metry is used the process is still dependent on camera calibration.
Considering the mobile robot mot ion, although low error values are present in the calcu lation of both distance and orientation, in the robot"s movement these values will be constantly corrected. The mobile robot captures images periodically and even if some measurements have some error.
By providing informat ion fro m the vision system, fu zzy logic mon itors the status of the robot. The robot is able to navigate in the laboratory, thus fuzzy logic controls the robot to move and position itself in the right conditions. The vision systems recognize the circle, the position of the circle and then align the robot to be in the center of the circle marking.
As the performance of an autonomous robot depend on the interaction of the robot with the real world for any mean ingful conclusion to be drawn, experiments have to be done to demonstrate that the robot can perform the desired tasks. Overall performance test of mobile robot was conducted to analyzed combination of fu zzy logic controller (FLC) ru les with desired speed.
The developed fuzzy logic control system co mbined with the vision system as a sensor to extract much information (target distance and target orientation) fro m the image to achieve a high level of self-adaptability is very important. The robot moved in d ifferent distances from the target and stop at 1.5 meter fro m the target because that is the minimu m distance that circle can recognize.
The Tables 4 and 5 show the performance of mobile robot in imp lementation of fuzzy logic controller. When the robot moving to the target, the results show that he average absolute error was about 0.03 meter and the average relative error was less than 3% for distance of robot to achieve it target. Given the measurement angle of orientation, it also revealed quite accurately with a mean absolute error of about 20•, and the average relative error of about 22%.

Conclusions
Land mark recognition is not an easy task but it is very important area in developing mobile robots that is need to accomplish positioning, path planning and navigation. A robot vision system using fuzzy logic technique to identify the object in real time was presented. The system is able to find and recognize the circle and then calculate the position of the robot to navigate in a laboratory. The technique used is based on image processing and optimization in real t ime. It hard to an able a mobile robot navigates in indoor environment. The output is very useful for a mobile robot in real t ime to find and detect objects. Usually a robot vision system is very slow due to computational time used for image p rocessing. Image processing time was min imized by the techniques used and the detection of door can be done about less than 1 second.
An analysis and design of fuzzy control law for steering control of the developed nonholonomic mob ile robot are presented. The proposed fuzzy controller is implemented on the developed mobile robot. The system can perform and satisfactory results are obtained which show that the proposed fuzzy controller can achieve the desired turn angle thus it can make the autonomous mobile robot moving to the target. The control system need PID controller to guarantee the stability of the straight conveying and turning trajectory.