AN ALGORITHM FOR OBSTACLE AVOIDANCE CONTROLLER USING ULTRASONIC SENSOR FOR MINI AIRCRAFT APPLICATIONS

Transcription

1 297 AN ALGORITHM FOR OBSTACLE AVOIDANCE CONTROLLER USING ULTRASONIC SENSOR FOR MINI AIRCRAFT APPLICATIONS Shridevi A Mali 1, G Sravanthi 2*, Siva Subba Rao P 3, Raja S 3,Sushma S J 1, A R Reddy 2, 3 1 GSSS Institute of Engineering &Technology for Women, Mysuru, India. 2 Madanapalle Institute of Technology &Science, Madanapalle, India. 3 CSIR-National Aerospace Laboratories, Bengaluru, India. ABSTRACT Ultrasonic sensor HC-SR04 was successfully used to measure the distance of an object. This sensor usually works on the principle of sensing the reflection of an impedance discontinuity. In-air types of sensors are used for the transition between air, solid or liquid surface. In the presence of many obstacles in front of ultrasonic sensor, nearer obstacle was detected initially.however, the ultrasonic sensor can be used for detecting the obstacle based on colour and size and hence collision can be avoided. In this paper an algorithm has been developed for avoiding collisions. This algorithm was embedded on an Arduino Uno that acts as control board. The quadcopter servo motors are controlled in the presence of any obstacle adjacent to the ultrasonic sensor. Initially the ultrasonic sensors were calibrated with different colours and plotted. Experiments were conducted on quadcopter and results were obtained. Results suggested that ultrasonic sensor detect the obstacle around 350cm than the other sensors as IR sensors and maxsonar. It is concluded that the algorithm implemented is useful in avoiding obstacle for mini aircraft applications as the control loop is faster in processing than other vision based algorithms. Keywords: Ultrasonic sensors, Quad copter, Obstacle avoidance, Distance measurement 1. INTRODUCTION Micro Aerial Vehicles (MAV) is nowadays built for hobby purpose, industrial or defence. The MAV are to be flown in any environment. The environment can be indoor or outdoor. Considering the environment the safety of the vehicle is important. Hence there were many research were done to maintain the safety of the vehicle like vision based technique by using cameras, laser ranging, IR sensors, sound sensors etc., were considered.the ultrasonic waves are sound waves that have the frequency above the limit of a human audibility. The ultrasonic sensor consists of a transmitter and receiver along with an ATmega16a microcontroller. From the microcontroller it transmits the pulse to the supply of the ultrasonic sensor and from the transmitter a pulse is triggered. Many different methods have been designed and developed for the measurement of distance from the observer to a target. This is done for navigation, surveying, determining focus in photography, accurately aiming a weapon, etc. Measuring the distance manually consists of lot of errors. IR rays were used for smaller distances. Thus, these disadvantages are eliminated by the use of ultrasonic sensor in the present work. 2. LITERATURE SURVEY Hear and avoid on MAV with acoustic vector sensors were considered in a paper [1]. The MEMS technology is used for this purpose. The 3D acoustic vector sensor consists of three orthogonally placed particle velocity sensor and one sound pressure microphone. For collision avoidance a depth map based on optical flow from on board camera images was obtained [2]. Here vision based technique was used. This approach is done to monitor the indoor corridor. Collisions are avoided by taking the real time images in a corridor with textured walls. An integrated system with a multimodal sensor setup for omnidirectional environment observation were implemented and it was equipped with a variety of sensors included a dual 3D laser scanner [4], three stereo camera pairs, an IMU and a powerful onboard computer to achieve the real time tasks. An autopilot used vision technology. Vision was combined with GPS and altitude sensors for waypoint navigation. Optic flow sensors were used and it was implemented on Kestrel autopilot version 2.2. this was equipped with three-axis accelerometers [5], absolute and differential pressure

2 sensors. A semi-autonomous MAV was designed at lower altitude. There were four stereo cameras connected on the quadrotor. This was optimized for on board computer vision. The images were reconstructed like structures and buildings. 3. ULTRASONIC SENSOR An ultrasonic sensor (HC SR04) was used to determine distance of an object. It offers excellent non contact range detection with high accuracy. It will detect from 2cm to 400 cm or 1 to 13 feet. Its operation is not affected by sunlight or black material. It has ultrasonic transmitter and receiver module. Fig1 shows the layout of the ultrasonic sensor which consist of pins like VCC, Trig, Echo and GND. The power supply required is 5V DC with a current 15mA. It has a resolution of 0.3cm and measuring angle of 30 degree. A trigger input pulse of 10μs is to be applied for the initiation of the sensor. The sensor in turn transmits out 8 cycle of ultrasonic burst at 40 KHz and need to wait for the reflected ultrasonic burst. When the sensor detects from the receiver it sets the echo pin to high and there is delay for period which is proportional to the distance. The distance is obtained by measuring the width echo pin. Time = Width of the echo pulse(μs) Distance (cm) = Time / 58 Distance (inch) = Time / 148 implemented on the MAV. The Arduino board is interfaced with the four ultrasonic sensors. The Sensors are positioned on the quadcopter on each side of the vehicle to cover 30 degree. The Arduino board is connected to the Arduino IDE the code is uploaded on the board. It is interfaced with the flight control board APM 2.6. MAV is powered with a 12VLiPo battery. This power is distributed to all the servo motors and by power module it will be supplied to the flight control board. The power from the flight control board is given to the microcontroller board and to all the ultrasonic sensors. The APM board is connected to many sensors like GPS, gyroscope, accelerometer. The servo motors are not directly connected to the board they are connected through Electronic Speed Controllers (ESC). The ESC controls the power to be given to the servo motors. The collected information is transmitted through the channel to the ground station from the telemetry. The baud rate is fixed through the mission plannar software. The ground station consists of telemetry and it receives at the same baud rate. The data of the MAV can be seen on the monitor using mission planner software. The RC transmitter is used to control the movement of the vehicle. The mission planner is used initially for calibrating the quadcopter. The radio calibration, compass calibration and accelerometer calibrations are performed. By using this software, the distance and the voltage of the ultrasonic sensor can be viewed and the vehicle behaviour for different colors can be observed. Fig 1: Layout of Ultrasonic sensor 298 Fig 2: Waveform of Ultrasonic sensor 8 cycle of 40kHz 4. BLOCK DIAGRAM The MAV consists of a transmitter part and a receiver part and in the ground to receive the data there is a transmitter and receiver part. The microcontroller board Arduino Uno is Fig 3:Block Diagram of the MAV 5. EXPERIMENTAL SETUP The initial calibration of ultrasonic sensor was conducted using quadcopter MAV it is shown in Fig:4. The sensors present on the APM board were calibrated before connecting to the quadcopter. Calibration like radio, compass, GPS and the ultrasonic sensors were done. The firmware was

3 loaded on the board depending on MAV we used. The ultrasonic sensor was connected to the control board through the analog pin. The sensor can be connected from the AN0 to AN5 pin here we have connected to the AN1 pin. The control board was connected to the interfacing software mission planner through USB cable. The initial settings for connecting to the ultrasonic sensor to the mission planner were done. A particular obstacle was taken to this all the colors were individually pasted and the corresponding voltage and distance shown on the mission planner and this was noted. The noted distance and voltages were plotted. Obstacle Distances Ultrasonic Sensor Quadcopter Fig 4: Initial Calibration of MAV The experimental setup consists of an MAV quadcopter as shown in Fig: 5(b). The quadcopter is implemented with all the sensors and control board at 8cm above the ground. It transmits the flight data to the mission planner. The RC transmitter is also connected to the flight control board and the baud rate is fixed. The servomotors are connected to the power module with the ESC to control the speed of the vehicle. In Fig:5 (a) an angle of 30 0 is measured in four direction is shown. The Arduino board is connected to the flightcontrol board. The trigger and echo pins of the ultrasonic sensors are connected to the Arduino Uno board. GPS Ultrasonic sensor Flight Control Board Arduino Propellers with servo (b) Fig 5:(a)Ultrasonic sensor mounted on MAV showing the angle at which it can detect. (b) MAV with different parts. 6. METHODOLOGY The MAV can detect obstacles and avoid collision by the following algorithm shown in Fig 6. The MAV is in flying condition it starts calculating the distance by ultrasonic sensor. The ultrasonic sensor starts continuously sending the pulses. The distances are collected by all the four sensors. When the distance collected by any of the sensor is less than 30cm then it is said that an obstacle is detected. If the distance obtained by the sensor is greater than 30cm, the MAV continues its mission. Then the sensor values are compared to know which sensor distance is less than 30cm. If only one sensor shows the distance value as lesser than 30cm then depending upon that pitch and roll movement is taken to move the vehicle in other direction as left/right/front/back after moving the vehicle away from the obstacle it continues its mission. If any two sensor value is shown less than 30cm then the vehicle has to be moved to other remaining direction by pitch and roll movement. This movement is performed by controlling the speed of the MAV using the ESCs. After the movement the mission continues and the distance are again obtained. 299 (a) Fig 6: Algorithm for Obstacle detection and avoidance

4 Orange (V) International Journal of Engineering Technology, Management and Applied Sciences If the distance obtained by three sensors are less than 30cm then the pitch and roll movement is performed in direction of the sensor showing larger distance which can be left/right/front/back. After the movement again the mission continues. If all the four sensors are showing the distance less than 30cm then the vehicle has to move in the upward direction by taking pitch or roll movement. After movement the vehicle continues its mission. The distances are again obtained from all the sensors. Hence the system detects the obstacle and avoids collision of the MAV in both indoor and outdoor environment. 7. RESULTS AND DISCUSSIONS The obstacle was placed from 2cm to 400cms and the respective reading was obtained in the mission plannar the obstacle considered was with different colors. There was a fluctuation in the voltage depending on the color of the obstacle used. For some of the colors the voltage became constant after particular distance. We checked for the colors in VIBGYOR and also white color. Fig 6 shows the results obtained by the mission planner for different colors. ultrasonic sensor displayed the distances and the required movement when obstacle was detected roll or pitch movement was also displayed in the plot. Fig 7: Serial Plotter for roll and pitch movement Fig:7 plots showed that when the obstacle detected the required pitch or roll movement is shown. In x- axis it showed the time and in y-axis it showed the roll and pitch movent values. These values represent MAV has to move towards left/ right/ front/back. In Fig:8 plots are obtained for the distance from Arduino IDE. In the plot x-axis represents the timer and y-axis represents the distance values of all four sensors. For different sensor different colors are used to show the distance Orange Black Green Red Yellow White Vilot Distance (m) Fig 6: Saturation of voltage for different colors. The voltage attains a saturation point for different colors. Fig:6 shows the graph obtained by the mission planner. The voltage varies with different distance. Different colors showed that the voltage attained saturation point at a particular distance. The MAV was connected with the telemetry and that was sending the results to the ground station through the telemetry which is been connected to the mission planner software. The code was interfaced with arduino and the distance was obtained in the serial monitor. All the Fig 8: Serial plotter for detecting distance of ultrasonic sensor 8. CONCLUSION In this paper tests were conducted for different color obstacle using Ultrasonic sensor (HC-SR04). The results show that the voltage was varying differently for different colors. The voltage became constant after some range, this indicates that for different colors the detecting range is different. The detecting range changes for each color with reference to the figure. The mission planner was used to take the voltage and distance values. The algorithm showed that the obstacle was detected and collision avoided. 300

5 ACKNOWLEDGMENT The authors would like to express thanks to the Director, CSIR-National Aerospace Laboratories for his endurance to carry out the work. Dr. Sathish Chandra, Head Structural Technologies Division STTD is bountifully thanked for providing all the facilities for completing this developmental work. Special thanks to Mr. Rajesh R, Project Assistant and Mr. Tiru Pathi, Project Assistant of Structural Technologies Division (STTD), CSIR-National Aerospace Laboratories, for experimental help. REFERENCES [1] Kirtan Gopal Panda*, Deepak Agrawal, Arcade Nshimiyimana, Ashraf Hossain, The effects of environment on accuracy of ultrasonic sensor operates in millimetre range, 20 February [2] Rajan P Thomas, Jithin K K, Hareesh K S, Habeeburahman C A, Jithin Abraham, Range Detection based on Ultrasonic Principle, February [3] Ultrasonic ranging module HC-SR04 datasheet [online] accessed on 23/08/ SR04b.pdf [4] H. He, and J. Liu, The design of ultrasonic distance measurement system based on S3C2410, Proceedings of the IEEE International Conference on Intelligent Computation Technology and Automation, Oct [5] Microcontrollers and Applications, Satish Shah, Oxford Publications. [6] ATmega16a data sheet, Atmel Corporation, U.S. [7] "Distance Measuring (Hurdle detection System) for Safe Environment in Vehicles through Ultrasonic Rays", Muqaddas Bin Tahir, Musarat Abdullah, Global Journal of researches in engineering, Automotive engineering, Volume 12 Issue 1 Version 1.0 February [8] "Accuracy and resolution of ultrasonic distance measurement with high-time-resolution crosscorrelation function obtained by single-bit signal processing" Shinnosuke Hirata1, Minoru Kuribayashi Kurosawa and Takashi Katagiri, Acoust. Sci. & Tech. 30, 6 (2009), The Acoustical Society of Japan. [9] E.H.G. Tijs, G.C.H.E. de Croon, J.W. Wind, Hearand-Avoid for Micro Air Vehicles Microflown Technologies, HAN University, [10] Simon Zingg, Davide Scaramuzza, Stephan Weiss, Roland Siegwart, MAV Navigation through Indoor Corridors Using Optical Flow, Autonomous Systems Lab ETH Zurich, [11] Marius Beul, Nicola Krombach, Yongfeng Zhong, David Droeschel, Matthias Nieuwenhuisen, and Sven Behnke, A High-performance MAV for Autonomous Navigation in Complex 3D Environments, International Conference on Unmanned Aircraft Systems(ICUAS),