UNIVERSITY OF OKLAHOMA GRADUATE COLLEGE Design and Analysis of a Four Wheeled Planetary Rover A Thesis SUBMITTED TO THE GRADUATE FACULTY in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE By Matthew J. Roman Norman, Oklahoma 2005 Design and Analysis of a Four Wheeled Planetary Rover A THESIS APPROVED FOR THE SCHOOL OF AEROSPACE & MECHANICAL ENGINEERING By Prof. David P. Miller Prof. Kuang-Hua Chang Prof. Dean Hougen c by Matthew J. Roman 2005 All Rights Reserved. Acknowledgements I would like to thank David Miller and the remaining faculty in the college of engi- neering for their helpful advice. They have directed the path I’m on toward a future that I dream of. Thanks to Malin Space Science Systems for providing the funds for this project. Thank you to all of my friends who have made sure that I learn from life outside the lab as well. Most importantly thanks to my family and their never-ending support. I could not have gone so far without such loving parents, Thank you Mom and Dad. iv Contents Acknowledgements iv List Of Tables viii List Of Figures ix Abstract xi 1 Introduction 1 1.1 Rovers for Exploration . 2 1.2 Rover Suspension Systems . 4 1.2.1 Independant Spring Suspension . 5 1.2.2 Articulated Body Suspension . 7 1.2.3 Rocker-Bogie Suspension . 10 1.2.4 Four Wheel Suspensions . 17 1.2.5 Legged Suspension . 18 1.3 Design Goals for Mars Rovers . 25 1.4 Terrain and Environment . 26 1.5 Four vs. Six wheels . 29 1.5.1 Are six wheels overkill? . 29 1.5.2 Why 4 wheels? . 32 1.6 Organization of Thesis . 34 2 Solar Rover-II Mechanical System 36 2.1 SR-II design goals . 37 2.2 Main Body . 38 v 2.3 Drive Train . 44 2.3.1 Wheel torque . 45 2.3.2 Mobility Power Requirements . 47 2.3.3 Motors . 49 2.3.4 Drive Train Concepts . 53 2.3.5 Motor Selection . 57 2.3.6 Power Transfer . 60 2.4 Suspension . 64 2.4.1 Central Differential . 64 2.4.2 Structure . 67 2.5 Wheels . 72 2.6 Electronics . 73 2.6.1 Sensors . 74 2.6.1.1 Obstacle Avoidance . 74 2.6.1.2 Tilt,Roll, and Heading . 75 2.6.1.3 Odometry . 75 2.6.2 Power . 76 2.6.2.1 Solar Panel . 76 2.6.2.2 Batteries . 76 2.7 Control System . 77 2.8 Construction . 78 3 Experimental Setup and Procedure 79 3.1 Rover Field Test . 80 3.1.1 Location and Terrain features . 81 3.1.2 Rover Setup . 81 3.1.3 Experiment Results . 83 3.2 Rover Laboratory Experiment . 87 3.2.1 Obstacle Traversing . 88 3.2.2 Slope . 90 3.2.3 Driving Power . 91 4 Results and Lessons Learned 93 vi Reference List 96 Appendix A Data Calculations . 103 Appendix B Mechanical Drawings . 117 vii List Of Tables 3.1 SR-II power used while maneuvering over various surfaces . 91 viii List Of Figures 1.1 Lunokhod, Russian for ”Moon Walker” (image reproduced from NASA) 4 1.2 Blue Rover and Robby are articulated body rovers designed by NASA(image reproduced from NASA) ......................... 7 1.3 Russian built Marsokhod (image reproduced from NASA) . 8 1.4 NASA’s Pathfinder rover on the 1997 mission and one of the twin Mars Exploration Rovers in 2004 (images reproduced from NASA) . 10 1.5 Link style mobility systems (images reproduced from NASA) . 12 1.6 Rocky Rover series (images reproduced from NASA) .......... 14 1.7 Changing direction ............................ 16 1.8 Sandia National Labs’ Ratler rover and Nomad rover (images repro- duced from [45] and NASA) ....................... 18 1.9 Ambler, a walking rover with a circulating gait and Dante, a frame walking rover (image reproduced from [8, 6]) .............. 19 1.10 Genghis and Attila biologically inspired hexapod robots (image repro- duced from MIT AI lab) ......................... 22 1.11 Rhex simplified leg design for a walking robot (image reproduced from [36]) .................................... 23 1.12 Qrio, a humanoid robot and Yambo-III a simplified biped robot (image reproduced from Sony corp. and [41]) .................. 24 1.13 Viking 2 landing site (image reproduced from NASA) ......... 27 1.14 Pathfinder landing site (image reproduced from NASA) . 28 1.15 MER Opportunity landing site (image reproduced from NASA) . 28 1.16 Sojourner climbing rocks (image reproduced from NASA) . 30 1.17 Four wheeled Solar Rover-II ....................... 33 ix 2.1 Solar Rover-II body and solar panel ................... 40 2.2 Honeycomb constructed body with the reinforced plate to which the geared differential housing is mounted .................. 42 2.3 wheel torque free body diagram ...................... 46 2.4 SR-II with motors in place ........................ 52 2.5 Belt drive with tensioning pulleys .................... 53 2.6 Chain and Sprocket drive with idler sprockets .............. 54 2.7 Drive train concepts using bevel gears and drive shafts ......... 57 2.8 Dual output bevel gear set and planetary drive ............. 60 2.9 Lower bevel gear set and wheel axle ................... 62 2.10 Cross section of SR-II’s hollow wheel axle ................ 63 2.11 Central gear differential mounted to the center of the body . 66 2.12 Tubular suspension structure ....................... 67 2.13 Central differential and motor mounting inside the body . 68 2.14 Upper gear box housing .......................... 69 2.15 Lower gear box housing (front) ...................... 71 2.16 Lower gear box housing (back) ...................... 71 2.17 Sharp infrared range finding sensor ................... 74 3.1 SR-II near the Salton Sea during the field test ............. 80 3.2 SR-II thermal delamination of the wheel ................ 85 3.3 SR-II position data taken during the field test ............. 86 3.4 Laboratory rover setup .......................... 87 3.5 SR-II climbing over a bump obstacle .................. 88 3.6 SR-II climbing over a step obstacle ................... 89 3.7 SR-II climbing a wooden plank slope ................... 90 3.8 SR-II outdoor slope test ......................... 91 x Abstract Rovers are important for conducting in-situ scientific analysis of objectives that are separated by many meters to tens of kilometers. Current mobility designs are complex, using many wheels or legs. They are open to mechanical failure caused by the harsh environment on Mars. This thesis describes Solar Rover-II, a four wheeled rover capable of traversing rough terrain using an efficient high degree of mobility suspension system. The primary mechanical feature of the SR-II design is its drive train simplicity, which is accomplished by using only two motors for mobility. Both motors are located inside the body where thermal variation is kept to a minimum, increasing reliability and efficiency. Four wheels are used because there are few obstacles on natural terrain that require both front wheels of the rover to climb simultaneously. A series of mobility experiments in the Southern California desert concluded that SR-II can achieve greater than 1km traverses in Mars like terrain during the six hours of peak solar energy per day. xi Chapter 1 Introduction Mobile robotic vehicles can be sent to an unknown surface and withstand the deadly environment of space with a much lower price tag and expenditure than a manned mission. The Russians landed two robotic vehicles on the moon and two more on Mars during the 1970’s, another three from NASA have landed on Mars since then. The rovers, Lunokhod 1 and 2 were able to explore regions further from the landing site and spend more time on the moon than a manned mission could have during that time. The two Russian Mars missions failed before achieving any science goals. In 1997 the Mars Pathfinder mission landed a small rover named Sojourner to explore the surface of the red planet. Two more rovers, Spirit and Opportunity, landed on opposite sides of Mars in January 2004 during the Mars Exploration Rover mission (MER). The Pathfinder and MER missions cost approximately $265 and $820 million 1 respectively, which is much cheaper than the $80 billion to $1 trillion estimates for landing a man on Mars. The rovers have the capability to conduct many science experiments in the area that they landed. Sojourner surveyed the area within about 10m radius around its lander, larger than the 3m reach of the arm on one side of the Mars Viking Landers. The MER rovers explored more than 5km away from their landers, which is equivalent to the average distance the lunar rover was driven away from the landing module during the Apollo program. These robotic missions have verified that remote science can be accomplished on the surface of another planet with a high degree of success. They allow access to areas of interest on the surface instead of being confined to the local area around the lander. 1.1 Rovers for Exploration The idea of sending a rover to the surface of another planet is to allow earth bound scientist’s access to specific areas of interest without enduring the harsh environ- ments of space [50]. The rover carries instruments to various terrestrial formations 2 for in-situ experimentation. The goal of the rover is to move between areas of inter- est quickly and safely. In order to better represent the planet of interest the rover must be able to travel tens of kilometers. Rovers designed for the exploration of other planets have had very complex mo- bility systems using large numbers of wheels or legs and sometimes multiple bodies. Two specific types of rovers have been to the surface of another planet: the Lunokhod rovers using an eight wheel design and three Mars rovers using the six wheel rocker bogie suspension.