Report
Four-
Legged Hexapod
Mentor –
Professor Vikram Kapila
Project
Leader–Isaac Osei
Name –
Dmytro Byvalets
Group
Members – Suchie Jaggi and Vincent Lopez
Date –
August 16, 2002
Abstract
The Robotics is the interaction of a combination between software and hardware with the outside world. The hexapod project is in the field of Robotics, where it has integrated the software and hardware to create a working, walking model. The robot has six operational feet, that will move it to the assigned destination. One of the advantages of walking robots is that they are better suited to handle rough terrain than mobile robots.
The experiment began with the building of a prototype and then integrating its design into the real robot. The servomotors that were used for the legs were calibrated and the Plexiglas pieces for its body were cut. The hexapod was able to move its body forward with the circular rotation of the two servomotors and the two aluminum plates. While the movement mechanism included only legs, a combination between legs and wheels would have produced better results.
Introduction
Mechatronics is the compatibility of hardware and software; it is the intersection of mechanical design and software design. In mechatronics, the software can be rewritten, this feature provides more freedom and cuts down on cost. In a project that is purely mechanical without the option of rewritable software, all the updates and repairs have to be done manually. For example, if a design is used in outer space and it brakes down unexpectedly, to make necessary repairs, it would have to be brought down to Earth. If the design were a mechatronics project, the changes could be made via a satellite.
The development of mechatronics, as the junction of mechanical engineering, is closely tied to the development of computers. As computers evolved from massive machines that could perform only simple tasks, into compatible complex everyday objects that are a vital part of society. As their price decreased, they began to gain new functions in the field of mechanical engineering. A computer is a vital part of mechatronics, because with it software is downloaded into "empty" hardware shells, that that project could use to perform whatever was requested of it.
In 1956, George Devol and Joseph Engelberger formed the world’s first robot company (ROVer Ranch). Five years later, in 1961 the first industrial robot, UNIMATE, was online in a General Motors automobile factory in New Jersey.
Large amounts of money have been invested in the development of robots. More efficient and multifunctional robots are developed every year. The robots are better suited for the hard, dangerous labor in factories because they are more efficient and they do not require a lot of maintenance and they have no salaries. There are two types of moving robots, mobile and walking robots. Mobile robots use wheels to transport themselves, and walking robots use feet, balancing their bodies while moving their feet forward. Walking robots are superior to mobile robots on rough terrain. Because a walking robot balances its body weight on its feet, when its legs aren’t aligned, it can still move forward. A mobile robot, with misalign wheel can turn over easily.
The first walking robots were interfaced with the big computers and so were also big. In 1970, the first mobile robot, Shakey, was built. This robot was able to move around the room, detect large blocks and then pick them up. Developments in mobile robots continued to be made; in 1994 Pionner I was the first robot to be available for sale at under $2,500. (Robot Ancestors)
The hexapod or six-legged robot has an advantage on rough terrain. Because the robot uses the PBASIC programming language as its software component, its functions can be rewritten and new functions can be assigned. The PBASIC program compiles data into binary numbers that are downloaded into the micro-controller. Because computers can only read binary numbers, this program makes it easier to download the program consisting of letters and words into the micro-controller. The hexapod robot performs movements by rotating an aluminum beam with the two legs in the circular motion using the servomotor. By alternating between the two sides and keeping the middle leg on the ground, it is able to move forward and maintain its balance.
Both mobile and walking robots have advantages and disadvantages. By combining the advantages of both designs, a better one could be created. The robot would use its legs on the rough terrain, and switch to wheels, for speed, on smooth ground.
The project will integrate the movement mechanism that is similar to the mechanism behind connection of the wheels on old-modeled trains. By positioning legs on the aluminum beam, the servo would only have to push that beam to move both legs forward, instead of moving each leg individually.
Methodology
The materials used for this project were: four 1.3 by .7 inches, two 2 by .5 inches, two 5.9 by .8 inches, four .5 by 3.2 inches, and two .5 by 3.0 inches aluminum plates. In addition two 4.0 by 7.6 inches Plexiglas pieces were used for the body of a walking robot and four 3.0 inch and two 2 inch bolts for its legs. The project was controlled by a Basic Stamp which was positioned at the top of the base 2 inches from the sides. Two servomotors were used for the movement of the four legs, one 3300 uF capacitor, and some wires, washers and nuts.
The first step in building this project was to create a base on which all the other parts will be positioned. For the base, piece of a transparent Plexiglas was used. Holes were drilled and then widened with a sander to fit the necessary pieces. Two holes were drilled for two metal plates which support and guide the beam that has the legs mounted on it. The first hole was positioned one .7 inches from the right side of the piece of Plexiglas and the other was 1.4 inches from the left end of the Plexiglas. The two openings are 2.4 inches away from the bottom of the Plexiglas piece.
The movement of the legs is controlled by a servomotor which rotates the beam with the two legs in a circular direction. The opening for the servomotor was created. It was 3 inches away from the right side of the Plexiglas piece and 2.1 inches away from its bottom. The opening itself was .8 by 1.8 inches and it fit the servomotor with its small propellers outside.
Since the middle legs of the hexapod would be stationary, two holes were drilled on both pieces of Plexiglas and then widened to fit those legs. That openings were each 3.8 inches away from the left side and 3.2 inches from the right side, and 1 inch above the bottom of the Plexiglas. The middle legs were thinner, approximately .1 inches in thickness, than the other four moving legs. For holding those legs in place two holes were drilled, the wider hole was .2 inches away from the end of the middle leg, and the other, a little narrower, was .8 inches from the end. In order to stabilize the legs and to raise the body, two metal bolts 2 inches long were placed in the openings on both the legs, .2 inches away from the opposite end. For each leg a nut was screwed in, then a washer was put on it and then the bolt was put through the opening and another nut was screwed in from the bottom to hold that bolt in place.
The four moving legs, two on each side of the Plexiglas, were attached to the 5.9 by .8 with the thickness of .1 inches aluminum beam. An aluminum beam was connected at the sides to the two smaller aluminum plates and in the middle to the servo propeller. The four legs were 3.2 inches in length, .5 inches in width and were .2 inches thick. To connect all four, two on each side of the Plexiglas, to the aluminum beam the same method was used as for connecting the middle legs to the pieces of Plexiglas. Two holes were drilled on one end, one .3 inches away from the end for a small bolt, and the other .8 inches from the end for the pin to hold it in place. On the opposite end of the legs, a hole was made for the 3-inch bolt, which had a nut and a washer on top of the aluminum leg, and another nut from beneath the leg to keep it in place.
The sides of the aluminum beam were connected to the metal pieces which in turn connected with the Plexiglas with the use of a bolt and three nuts. Two of the nuts were placed in between the two aluminum plates, to preserve the distance between the plates, and then another was placed on the outside of the beam.
To hold the Basic Stamp on top of the Plexiglas, two openings were created for the two small Plexiglas pieces, approximately 3.8 inches in length, that stretched from one side of a Plexiglas body piece to the other. Then the Basic Stamp was connected to those two Plexiglas pieces with the four bolts on the four corners of the Basic Stamp.
The two servos were
connected to the stamp, with middle wires connecting to the positive end of the
3300 uF capacitor, black side wires going to the VSS and the other two color
wires going to the P13 and P14 channels. The negative side of the 3300 uF
capacitor was connected to the VDD power source. The program for the hexapod
was written in PBasic, and then downloaded onto the Basic Stamps with the
connection cable.
Results
The Four-legged Hexapod was built over the course of the seven weeks allotted to that project. The hexapod combined two servomotors and a combination of Plexiglas and aluminum plates for its body to create a walking robot. The two servomotors were calibrated and their numbers for which they stop rotating were recorded. The servo with the white wire connected to the P13 channel had a middle number of 340. The other servomotor, connected to the P14 channel had a 1225 for its middle number.
The final model of hexapod was built upon the redesigns made on the previous prototypes. (Figure 1) The first prototype was built from Lego Robotics system kit, the next redesign lead to the creation of the new aluminum body for the hexapod. The errors from those two prototypes were corrected in the third model of a hexapod.
The servomotors were connected to the Basic Stamp. (Figure 2) The hexapod model performed movement by rotating its two feet on the aluminum beam in the circular motion with the use of the two servomotors. (Figure 3)
Figure 1: First Prototype
Figure
2: Servomotor Schematic
Figure
3: Final Hexapod Model
Discussion
The final hexapod project turned out completely different from the first planned model. The first model was going to integrate the sensors and video camera along with the leg mechanism that would allow it to make turns based on readings from the sensors. That first model was planned for fire situations. During any kind of fire hazard, that robot would go in and transport the readings from its sensors along with the camera pictures regarding any kind of survivors present in that dangerous place.
The first problem that was encountered was in the leg mechanism. Because the middle leg, when working, turned in the opposite direction as the other two that were connected to the same servo, the design had to be corrected. In the new design, the middle leg was connected to the first big gear on the left and with this connection it functioned in perfect synchrony with the other two front legs.
Once the Lego prototype was finished, and all design errors were corrected, a new software error emerged: the micro-controller that was used on the prototype was incompatible with the Lego software that was installed on the computer.
Because of their incompatibility there was no way to test the programming for that prototype. The project took another step after this software error. For the new design, to lighten the load on the legs of the hexapod, aluminum plates were used for the body instead of the Lego pieces. It also had two real servomotors, and a real micro-controller Stamp. For the connection of the three legs to one servomotor, a similar technique of Lego gears and Lego rods was used. The hexapod was put together and the program was written in PBasic. When that design was tested with the new modified program, new errors arose. The hexapod was too heavy for the plastic Lego rods to support, and when the weight would fall on them, they would bend. The middle leg had no support for holding the weight of the hexapod’s body, and the two servomotors would not move at the same time and at the same speed.
The final redesign for the hexapod project integrated a whole new leg movement system. It has legs positioned on the aluminum beam, which is moved by the motion of the servo, and the middle legs on both sides stayed firmly on the ground to provide balancing support for the moving side of the hexapod.
The final design solved many previous problems that the earlier designs had to face. It doesn’t use plastic rods for the leg connection, and does not use unreliable plastic Lego gear for its mechanism. Instead, the two servos are directly connected to the two moving legs, on each side, by the aluminum beam. Because the design does not use gears to move the legs, the servos do not loose momentum in their rotations and do not waste unnecessary energy on moving the gears.
The final problem that the hexapod faced was that the servos received small amount of power when running. During testing, the servos worked fine alone, but when they were made to follow one another, their power greatly decreased. This problem was solved by connecting a positive end of a 3300 uF capacitor to the red wires of the Servos and then connecting the negative end of the capacitor to the VDD.
Conclusion
The four-legged hexapod has met the objectives that were set out at the beginning of its construction. The hexapod was able to perform the main function that was required of it; it was able to move its feet forward while maintaining its balance on the remaining three feet that stayed on the ground. By creating different prototypes, several major errors were found in their designs, the discovery of those errors led to the creation of the final design which corrected them.
Because of the limited
amount of time set to this project, not all of the planned components were used
within the hexapod’s system. The use of wheels along with the legs was left out
from the designs. In the future research it might have been enlightening to use
the combination of wheels and legs. In addition more advanced sensors with
higher sensibility might have been used to improve the hexapod’s movement. The
sensors such as photodiode, motion and infrared along with piezo speaker for
sound and voice recognition program could be integrated into the working
walking hexapod’s system. In addition, a muscle wire could have been used for
the movement of the two middle legs.
References
"Quick Facts." Globalization: The Story Behind the Numbers <http:www.imf.org/external/pubs/ft/fandd/2002/03/picture.html> (July 7, 2002)
Computer use and Ownership Data. <http://www.census.gov/population/socdemo/computer/p23-155/tab01.pdf> (July 09, 2002)
Jason Beard et al. , "Computer Chronicles: From Stone to Silicone." <http://library.thinkquest.org/22522/timeline2_en.html> (July 07, 2002)
Alciator D. and Histard M., Introduction to Mechatronics and Measurement Systems, McGaw Hill, 1999
"Robot Timeline", A short History of Robots. <http://prime.jsc.nasa.gov/ROV/history.html> (July 17, 2002)
"PBasic Compiler", PBasic Information. May 1996, <http://www.labyrinth.net.au/donmck/more.html> (July 17, 2002)
"History of Mobile Robots", Robot Ancestors. 2000, <http://www.activrobots.com/HISTORY> (July 21, 2002)
"Development of a walking machine: mechanical design and control problems", Mechatronics. June 2002, <http://www.sciencedirect.com/science/journal/09574158> (July 21, 2002)
Hogan James, Exploring the world of Artificial intelligence: Mind Matters, New York: The Ballantine Publishing Group, 1997
Moravec Hans, Robot: Mere Machines to Transcendent Mind, New York: Oxford University Press, 1999
Acknowledgments
Lab Supervisor – Sang-Hoon Lee
Mentor – Professor Vikram Kapila
Teaching Assistant – Isaac Osei
YES Center
And
Polytechnic University for making this program possible.