CRANFIELD UNIVERSITY Anthony Dumisani Masuku PLANETARY ROVERS AND DATA FUSION SCHOOL OF ENGINEERING Astronautics and Space Engineering MSc THESIS CRANFIELD UNIVERSITY SCHOOL OF ENGINEERING MSc THESIS Academic Year 2013-2014 Anthony Masuku Planetary Rovers and Data Fusion Supervisor: Dr. Stephen Hobbs May 2012 This thesis is submitted in fulfillment for the degree of MSc by Research (cid:13)c Cranfield University 2013. All rights reserved. No part of this publication may be reproduced without the written permission of the copyright owner. i Abstract Thisresearchwillinvestigatetheproblemofpositionestimationforplanetaryrovers. Diverse algorithmic filters are available for collecting input data and transforming that data to useful information for the purpose of position estimation process. The terrain has sandy soil which might cause slipping of the robot, and small stones and pebbles which can affect trajectory. The Kalman Filter, a state estimation algorithm was used for fusing the sensor data to improve the position measurement of the rover. For the rover application the locomotion and errors accumulated by the rover is compensated by the Kalman Filter. The movement of a rover in a rough terrain is challenging especially with limited sensors to tackle the problem. Thus, an initiative was taken to test drive the rover during the field trial and expose the mobile platform to hard ground and soft ground(sand). It was found that the LSV system produced speckle image and values which proved invaluable for further research and for the implementation of data fusion. During the field trial,It was also discovered that in a flat hard surface the problem of the steering rover is minimal. However, when the rover was under the influence of soft sand the rover tended to drift away and struggled to navigate. Thisresearchintroducedthelaserspecklevelocimetryasanalternativeforodometric measurement. LSVdatawasgatheredduringthefieldtrialtofurthersimulateunder MATLAB, which is a computational/mathematical programming software used for the simulation of the rover trajectory. The wheel encoders came with associated errors during the position measurement process. This was observed during the earlier field trials too. It was also discovered that the Laser Speckle Velocimetry measurement was able to measure accurately the position measurement but at the same time sensitivity of the optics produced noise which needed to be addressed as error problem. Though the rough terrain is found in Mars, this paper is applicable to a terrestrial robot on Earth. There are regions in Earth which have rough terrains and regions which are hard to measure with encoders. This is especially true concerning icy places like Antartica, Greenland and others. The proposed implementation for the development of the locomotion system is to model a system for the position estimation through the use of simulation and col- ii lecting data using the LSV. Two simulations are performed, one is the differential drive of a two wheel robot and the second involves the fusion of the differential drive robot data and the LSV data collected from the rover testbed. The results have been positive. The expected contributions from the research work includes a design of a LSV system to aid the locomotion measurement system. Simulation results show the effect of different sensors and velocity of the robot. The kalman filter improves the position estimation process. KEYWORDS:DataFusion,planetaryrover,KalmanFilter,navigation,LaserSpeckle Velocimetry, measurements, Mars. iii iv Acknowledgements I am sincerely and heartily grateful to my supervisor ,Dr Stephen Hobbs , for the support and guidance he showed me throughout my study and thesis writing. I am sure it would have not been possible without his help. Besides I would like to thank my mother and my friends who boosted me morally and provided me with great information resources. A thanks goes out to Jayden Nkosikhona Masuku for inspiration. Furthermore, I would like to thank the following: Chimu - Advice about MSc study at Cranfield Dr Daniels (Engineering Photonics) - Assistance and advice (Laser Speckle Ve- locimetry) Boris Snapair - Technical support with Matlab Dr Charrot (Engineering Photonics) - Collaborating in this project (Laser Speckle Velocimetry and the supply of measurement data. Heather Woodfield (Social Sciences Information Specialist) - Refworks and Write n Cite training and for being kind. Alison and Deborah(IT Training and Support Specialist) - Support using MSWord and thesis formatting Mrs Millia Masuku - Support and encouragement from beginning to end. Having lots of faith in me. Fuller Masuku in remembrance. My Tribe - Zwide ka Langa, Nhlane, Gudunkomo,Zikode, Ndwandwe, Mkhatshwa, Nxumalo, Tunda bathola bafokazane. A Psalm of David. 23 The LORD is my shepherd; I shall not want.2 He makes me to lie down in green pastures; He leads me beside the still waters.3 He restores my soul; He leads me in the paths of righteousness For His names sake.4 Yea, though I walk through the valley of the shadow of death, I will fear no evil; For You are with me; Your rod and Your staff, they comfort me.5 You prepare a table before me in the presence of my enemies; You anoint my head with oil; My cup runs over.6 Surely goodness and mercy shall follow meal the days of my life; And I will dwell[a] in the house of the LORD Forever. . v vi CONTENTS Contents Contents vi List of figures x List of tables xii Abbreviations xiii 1 Introduction 1 1.1 Robotics Missions Background . . . . . . . . . . . . . . . . . . . . . . 2 1.1.1 Aims . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1.1.2 Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 1.2 Dissertation Structure . . . . . . . . . . . . . . . . . . . . . . . . . . 4 2 Literature Review 5 2.1 Planetary Rovers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 2.1.1 On board Sensor . . . . . . . . . . . . . . . . . . . . . . . . . 6 2.2 Locomotion Subsystems . . . . . . . . . . . . . . . . . . . . . . . . . 8 2.2.1 Rover Navigation . . . . . . . . . . . . . . . . . . . . . . . . . 9 2.2.2 Wheel Bogie System . . . . . . . . . . . . . . . . . . . . . . . 11 2.2.3 Simulation of a two wheel rover . . . . . . . . . . . . . . . . . 13 2.3 Optics and Measurements . . . . . . . . . . . . . . . . . . . . . . . . 13 2.4 Speckle Velocimetry . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14