Submarine Fishing Assistant

Transcription

1 TVE-E Examensarbete 15 hp Juni 2018 Submarine Fishing Assistant Independent Project in Electrical Engineering Joel Larsson David Forsberg

2 Abstract Submarine Fishing Assistant Joel Larsson, David Forsberg Teknisk- naturvetenskaplig fakultet UTH-enheten Besöksadress: Ångströmlaboratoriet Lägerhyddsvägen 1 Hus 4, Plan 0 Postadress: Box Uppsala Telefon: Telefax: Hemsida: Humanity has been fishing for a long time, more specifically under ice with nets, this could be a lot more efficient if it could be automated in any stage of the procedure. The project aim is to construct an automated system for assisting in net fishing under ice. A navigation system using a GPS module, an IMU and radio transceiver was designed, constructed and implemented on a car for testing. The result was a system that could navigate to either a predetermined point or return to a point saved by the system. A "manual mode" was also implemented to be able to operate the system as a remote controlled car with a computer coupled to an Arduino with a radio transceiver sending control signals to the prototype. The prototype worked as intended, but would need further improvements to the automated navigation system's precision to be ready for commercial use. This should be the primary focus of any further development of this project. Handledare: Kjell Staffas Ämnesgranskare: Ladislav Bardos Examinator: Hana Barankova ISSN: , TVE-E

3 Contents 1 Vocabulary 2 2 Introduction Background Objective Limitations Theory GPS DC motor driving IMU Accelerometer Gyroscope Magnetometer Experimental Details Planning Process Mechanical Construction Electronic Assembly Verification of the system Programming Results Mechanical Construction Sensor Fusion The prototype system as a whole Discussion Precision of product Practicality of product Other applications Conclusion Acknowledgements

4 1. Vocabulary IMU - Inertial Measurement Unit I2C - Inter-Integrated Circuit; an electronic communication system DC motor - Direct Current motor GPS - Global Positioning System Arduino (IDE) - Arduino (Integrated Development Environment) CAD - Computer Aided design, software used for designing constructions in 2D and 3D. 2

5 2. Introduction 2.1 Background Ever since mankind realized there are huge quantities of food roaming the seas, fishing has been a central part of survival. This has lead to multiple technical advancements in the naval and fishing field. Ice fishing has always been a big part of the fishing culture in the colder parts of the world, and this has traditionally been done in a primitive manner. Dragging a net between two bored holes in the ice to catch fish is common practice in Sweden. An ice jigger is commonly used in this scenario, to first drag a line between the bored ice holes, and the line is in turn used to mount the fishing net under the ice. One could imagine this work to be cumbersome, therefore any kind of technical advancement that could alleviate the working process would be of great assistance. 2.2 Objective The objective is to design a product that can aid in the process of laying net under ice. The process that this project aims to facilitate is dragging a line between two holes in the ice. The vehicle should if possible be able to navigate between the holes by itself and travel a distance of up to 70 meters under the ice. The vehicle should also function if the ice is bumpy and be able to salvage if it gets stuck under the ice. The specifications of the prototype are that it should be able to navigate to a point with a precision of 2 decimeters since the hole has a diameter of about 4 decimeters. It should also be able to travel the 70 meters it is supposed to do under the ice. The battery should be large enough to last 3 of these 70 meter trips. 2.3 Limitations Due to the potential complexity of building and testing a submarine capable of navigating between bored ice holes, the project naturally needed some limitations. Main focus of the project work lied in designing and producing a highly precise and safe automated control/navigation system of the submarine, using a GPS-module and a IMU sensor coupled with an Arduino microprocessor. Because of the time frame of the project the submarine construction was only designed and never constructed nor tested in practice, instead to verify the system a RC-car format was used. This format ensured that the system could be verified and tested in a reasonable time frame. 3

6 3. Theory 3.1 GPS GPS is a global navigation satellite system maintained by USA. The GPS system functions by measuring the distance between satellites with known position in space and a receiver with unknown position. A minimum of four satellites is required to determine all parameters needed, a three dimensional coordinate and the error in the receiver clock. The distance measurement can be done in two ways, by code and carrier phase measurement. Code measurement is done by measuring a time shift between code received and code generated in the receiver. The code is generated at exactly the same time in the satellite and the receiver so by multiplying the time delay by the speed of light the distance can be calculated. Carrier phase measurement is done by measuring the number of wavelengths between the satellite and the receiver. The fraction of a wavelength is determined by measuring the phase shift with respect to the generated signal in the receiver but the number of whole wavelengths is more difficult to determine as the carrier signal has no timestamps. Because the carrier phase measurement is more complex it takes a longer time to get a valid position, therefore it is often complemented by code and differential measurement. The advantage of phase measurement is that a higher accuracy can be achieved, in theory phase measurement can be done with an accuracy of about 2 millimeters while code measurement has an accuracy of about 3 meters. Some sources of errors are faults in the satellite trajectory, distortion in the ionosphere due to charged particles and multipath errors. The error due to the satellite trajectory is caused by the satellite position not being the same as the broadcast one. The effect can be reduced by doing relative measurements or by correcting the data with the actual satellite position later. Distortion in the ionosphere is caused by the signal interacting with the charged particles and the effect is proportional to the frequency. It can be reduced by doing relative measurements or measuring signals on different frequencies. Multipath errors occurs when the signal is reflected of a surface before reaching the receiver therefore giving the wrong distance. This kind of error is more common when tall buildings or trees are close by. By measuring over a longer period of time and using good equipment this error can be reduced.[1] 3.2 DC motor driving In order to be able to control a DC motor to turn both directions, the current has to be able to flow both directions. Therefore to control the current bidirectionally through the motor, an h-bridge has to be implemented in the system. An h-bridge consists of 4 MOSFETs, 2 NPN and 2 PNP that are excited in a way that lets the current flow in one direction but not the other. 3.3 IMU An IMU is a sensor that can measure the force, angular movement and sometimes electrical field it is subjected to. They are often used in aircrafts, both manned and unmanned, to determine the orientation of 4

7 the plane during flight Accelerometer The most popular kind of accelerometer is based on the piezoelectric effect but other types are also used when the acceleration have a low frequency. They operate by having a mass suspended by a piezoelectric material acting as a spring. When the sensor is subjected to an acceleration the mass will cause stress on the piezoelectric material and generate an electrical signal. An IMU usually have three accelerometers oriented orthogonally to be able to detect acceleration in all directions.[2] Gyroscope Gyroscopes in IMUs are often angular rate sensors which measures change in the angular rotation. It works by oscillating a structure that due to the Coriolis effect will get a secondary motion in another direction when turned around the measurement axis. By measuring the amplitude of the vibrations in the secondary axis the rate of the rotation can be determined. There is usually three gyroscopes oriented orthogonally in an IMU to be able to detect rotation in all directions.[3] Magnetometer One class of magnetometers is Hall effect sensors that measures the voltage caused by the Hall effect across a conductor. When a current is flowing through a magnetic field perpendicular to the conductor the electrons will experience a force that will deflect them to the right or left depending on the direction of the magnetic field. By measuring the voltage across the conductor the magnetic field can be determined. It is also possible to measure the change of resistance as a result of the changed path of the electrons. By arranging three magnetometers orthogonally the magnetic field can be measured from all directions.[2] 5

8 4. Experimental Details 4.1 Planning Process The first step of the planning process was trying to find suitable components for the navigation/control system, naturally similar problems and their solutions were looked at. Similar projects involving GPSmodules and absolute orientation sensors have been done before, therefore there was a lot of research involved at this stage trying to find the most optimal solution for the task at hand. The choice of using Arduino in conjunction with accompanying modules came with a good interface and practical implementation of programming. 4.2 Mechanical Construction An old experimental segway robot was used for a lot of its parts such as; battery pack, DC motors, motor driver card, tires and breadboards. The new parts such as Arduino Uno, IMU, GPS-module and GPS antenna were added to the construction. The first testing prototype can be seen in figure 4.1. Figure 4.1: The prototype before a custom made chassis was constructed. The on-land car prototype was assembled together on a new custom made chassis made out of plexiglass 6

9 to make the layout of the components more suitable. A radio transceiver was added to the design to be able to control and monitor the vehicle remotely. The second radio transceiver was connected to another Arduino which in turn was connected to a computer, using the Serial monitor in the Arduino IDE as interface for communication. A very basic concept of the submarine design seen in figure 4.2 was done in CAD to visualize the final product. "Skates" were mounted on the top of the submarine to let it skate beneath the ice. Figure 4.2: CAD of the submarine, the holes are prepared for the propeller axes. 4.3 Electronic Assembly Most of the electronic assembly involved getting the sensor fusion to work properly, only one microprocessor was used on the vehicle itself. The Arduino Uno had just enough connection pins so it could connect to the IMU, GPS-module, motor driving circuit and the radio transceiver. To be able to drive the DC engine bidirectionally an L298 motor driver circuit was incorporated. The IMU was positioned in the front of the vehicle as far away as possible from the battery pack and engine which are located on the rear axis, this is an attempt to minimize the impact of statics from engines on the measurements of the IMU. BNO055 by Adafruit is the IMU sensor used in this build, it s an absolute orientation sensor with a 3- axis accelerometer, 3-axis gyroscope and a magnetometer. It uses the I2C protocol to communicate with other components. RoyalTek REB-4216 is the GPS-module used mainly because it was mounted on an Arduino shield system that is made to interface with an Arduino Uno, this was preferable since any kind of ease of implementation is helpful in a project of this complexity. 7

10 4.3.1 Verification of the system Tests were conducted on the robot to try and verify the system. It was programmed so it could store it s current position, then use these coordinates as it s next destination. Location of these tests had to be conducted as far away from a building as possible to minimize the effects of multipathing from the satellite signal but also on adequate terrain for the robot. 30 meters southwest of the Ångstrom laboratory was used as a frequent testing site. First off the BNO055 had to be calibrated to get it s offset, clear skies is also preferable to get a more reliable satellite signal and also enough satellites had to be in contact with the GPS module. It would then be put in place where the destination was wished to be set, store the coordinates for the current position and then move it 70 meters away and let the automated navigation system take it to the coordinates that were stored. The robot would stop when the distance to the destination was less than a meter and send confirmation of this to the computer via radio signals. Measurements were then made of the physical distance from the point of the stored coordinates to where the robot had arrived at its destination. This test was conducted dozens of times. 4.4 Programming The microprocessor used was an Arduino Uno R3. Programming was exclusively done in Arduino IDE, this is open source software that is based on the programming language Java and it can utilize both C and C++. Initially the parts got individual programming and testing to verify that they fulfilled the requirements of the project. The GPS-module is used to retrieve the current coordinates of the vehicle, these can then be used to calculate the course to the destination coordinates (that can either be preset or retrieved locally using the GPS-module). Then the magnetometer in conjunction with the accelerometer are used to align the vehicles course with the course to the destination, turning and propelling itself using the DC motors. A flow chart of the system can be seen in figure

11 Figure 4.3: A block diagram of the system. 9

12 5. Results 5.1 Mechanical Construction The prototype build turned out to be durable and reliable in regards to driving on flat terrain. This made testing of the prototype ideal in urban areas. Due to limitations mentioned earlier in the report, the submerge able vehicle was never constructed, only designed. This car prototype was never intended to be submerged. 5.2 Sensor Fusion Using multiple sensors with a micro processor to increase performance seems to have been a success. The software library TinyGPS++ by Mikal Hart was implemented into the code, this granted a lot of functionality and pre-made functions regarding the GPS module. The BNO055 absolute orientation sensor handles calculations of raw measurement data from the IMU on the chip, calibrating the sensor and giving output in the form of Euler angles. Software library Adafruit_BNO055 by Kevin (KTOWN) Townsend for Adafruit Industries was implemented into the code to increase functionality with pre-made the sensor. 5.3 The prototype system as a whole It came together as imagined. The prototype could navigate by itself from point A to point B. The system can easily be switched over to "manual mode" and remote controlled by a computer using another Arduino Uno microprocessor coupled with a computer and a radio transceiver. From the tests described in it is clear however that the precision of the automated navigation system is roughly ±2 m. The specification that the battery would last at least 3 trips is more than fulfilled since the battery was only charged once during all the tests. 10

13 Figure 5.1: The final version of the automated control system prototype. 11

14 6. Discussion 6.1 Precision of product Tests showed that the precision of the system was around ±2 m. A bored hole in the ice used for ice fishing is usually around four decimeter in diameter. That means that the automated navigation system isn t precise enough, it s even more likely that the GPS module has worse performance under ice due to the signal being subject to refraction and multipathing in the ice. This is something that would be prioritized for improvement if there was more time on the project. To increase the precision of the system one could possibly implement one of the following theoretical systems: Base GPS reference station; if the exact coordinates of a fix GPS station is known, it s possible that it s current position being calculated from satellites is a little bit off, since the exact coordinates of the base station is known this could be measured. The idea here is the same offset is likely applicable to nearby GPS for instance the car prototype, if so it would increase the precision of the system to acceptable levels. Ultrasound sensors and speaker; Two ultrasound sensors could be mounted on the front of the submarine and listen for a speaker submerged in the bored ice hole, sending out ultrasound waves. Then by measuring the phase shift between the two captured signals in the sensors the system could in turn be able to tell in which direction the bored ice hole is and align it s course appropriately. 6.2 Practicality of product The goal of the project was to design a product to aid in the process of getting a line between two holes in the ice. To be a good alternative to the methods traditionally used the product needs to be reliable. The prototype as it is now have not shown the level of reliability to completely replace traditional methods. Aside from the need to increase the precision the calibration of the IMU could be improved. As it is now the IMU can lose its calibration and would need to be recalibrated before the submarine can continue to drive. To recalibrate the IMU it needs to be moved in random figure eights until it indicates being calibrated. As it is impossible to move the sensor in figure eights under the ice another way to solve this is to save the offset data from the sensor and upload it back to the sensor when it loses calibration, assuming the environment around the submarine has not changed to much. The problem with doing this is that the sensor still have a status of not being calibrated when the offsets have been uploaded. The sensor continuously calibrates itself and it might be necessary to keep uploading the offsets to ensure that the right values are kept. Since the sensor is like a "black box" in regards to the calibration and calculation software it makes it harder to implement in the design. 12

15 6.3 Other applications The submarine could probably be used in other applications not being limited to only operate under ice. It could easily be modified to drive towards either a predefined coordinate or transmitter making smaller deliveries at sea when it is not safe or practical to do by boat. The navigation system could also be used in other systems where a driver is not needed, like ocean cleaning. 13

16 7. Conclusion The submarine automated control and navigation system worked as intended, and would be ready to implement into the submarine design. However the developed system would probably need further work to be ready for commercial use. In particular the precision of the system should be improved and could be a subject for future research. The system could also be modified to have a wider range of applications as new ways to use drones are invented constantly. 7.1 Acknowledgements The department of electricity at Uppsala university has been a great help by funding this project and providing equipment, thank you. We would also like to thank Kjell Staffas for being the supervisor of this project, he provided us with great insights in the technical- and project field which in the end affected the end products outcome greatly. 14

17 Bibliography [1] Lantmäteriet, GPS och satellitpositionering se/sv/kartor-och-geografisk-information/gps-och-geodetisk-matning/ GPS-och-satellitpositionering/ [2] Wilson, Jon S, Sensor technology handbook, 2005, Elsevier [3] Apostolyuk, Vladislav, Coriolis Vibratory Gyroscopes: Theory and Design, 2016, Springer International Publishing 15