This week, we worked hard on the construction of our final iteration. First and foremost, we printed out the main body of our walker and assembled it using wood glue. In this final iteration, we also made a removable hatch that would be used to hide and protect the electronics located inside of the robot. Then, we moved onto the design of the legs and of the feet of our robot. Though the printing process was relatively easy, we had to adjust certain parameters of the walking mechanism and of the legs. One of the major issue we ran into was the fact that one of the connection shaft had to be bent in order for the wheel to fully complete a 360 degrees rotation. In order to solve this issue, we added a lengthened socket on one of the links of the walking mechanism. Furthermore, we also had to lock the motors onto the walls of the inside of the craft. While we could not find suitable nuts for the type of screw we needed (until after the problem was solved... ;( ), we used lock nuts to replace conventional nuts to attach the motors to the robot. In addition, we also had to cut out numerous shafts for the connections of the legs. As many of the connections required an unique length to fit in the leg, it was a perilous task to cut out and organize them. There were also a large amount of work required in preventing the different links of the legs and of the walking mechanism from colliding with each other. Despite the difficulties, we managed to create a specific layout that prevent the interference between each linkage. Next week, we plan in finishing the final iteration. We would continue working on the connections and we would test out the effectiveness at walking of our robot.
Today, we searched for a suitable design for our project. In the process of doing so, we found several interesting designs which could be used in our project. For instance, one of the precedents we deemed interesting involved a triangular leg design. It had many advantages, such as the ability to jump, climb stairs and to move with a relatively high speed. Despite the benefits of this design, however, we had trouble figuring out the way how it works and how to build it. A second design we found consisted of a legged robot concept which resembled a cheetah. It could walk over obstacles without difficulty and could reach a high walking speed. Moreover, due to its simplier design, we were able to understand the way how it works and we had an idea in how to build it. During the process of our brainstorming, we decided to use pistons to power the legs, for they were simple to design and make compared to other methods. Next class, we plan to complete our sketching process and to start experimenting with different designs of legged walkers.
Last week, due to Hurricane Irma, we only had one period of Innovation Studio. As a result, we were not able to gain much progress in our main project. Instead, during last Thursday's class, we had a short tutorial on how the Arduino works and how it can be used. For instance, we learnt how to make a LED lamp flash by uploading a specific Arduino code onto Arduino and by connecting the LED light to the right emplacement. Moreover, we also learnt how to connect servos to the Arduino and how to operate them using either a potentiometer or an Arduino code. Finally, we also played around with the settings of the Arduino code in order to find out how different parameters would affect the behaviour of the LED light and of the servos. In the next class period, we plan to resume the construction of our prototype.
This week, we continued the designing and construction of the prototype for our legged walker. First, we completed a whole mockup of the robot using cardboard. We built this simple model in order to have a better understanding of the placement of the legs and of the different electronic components. Then, we started working on a larger and more detailed prototype of the legs for the walker. At first, we only assembled the front legs of the walker. However, we soon realized that a different design of the rear legs was required to ensure the proper performance of our creation. As a result, we also worked on the design of the rear legs and we managed to complete it before the end of the week. During the our designing process, we also worked on the method of propulsion for our legged walker. At first, we decided to use servos to power the legs of the robot. As we progressed in our designing process, however, we found out that in order to reproduce a fluid and realistic walking motion, there would have to be more than one power source for each leg. Hence, we decided to abandon the servo design and use motors instead. (Additionally, one of the advantages that came with the use of motors was that it would be possible to power two gears with only one motor by linking them with a transmission system.) Next week, we plan in using Fusion to design and test out the propulsion system for the legs of the walker.
Today, we gave a presentation on the current progress of our project and received feedback on the current stage of our legged walker. According to the nuvu coaches, the rate at which our process of designing is good. Furthermore, we received positive feedback on our initial sketches and ideas. After our presentation, we were given a numbers of advices for what should be done in the near future. For instance, it was proposed that we should not linger with simulation of gears in fusion 360 and that we should print out the whole mechanism as soon as possible. In addition, it was proposed that we should start with building only one leg and make the final product slightly lower and wider.
In future periods, I intent to apply these propositions. Depending on the time left for our project, we would make two separate leg designs for the front and rear leg or we would only copy and paste one single leg design for all four legs of our walker. Overall, today’s nuvu class was very useful in giving us a clearer vision of the future of our project.
With the feedback we received from last week’s presentations, we decided it was time to print out the walking mechanism for our robot and to test it out. First, we had to choose between several concepts of walking mechanism. One of them, for instance, relied on only one connection to the powered wheel, while another design proposed to have two points of attachments with one connected to the powered wheel and the other rooted onto the craft itself. After some reflection over the matter, we decided that it would be more adequate to use the latter design. We modeled it in Rhinos and cut it out of wood using the laser cutter. Then, we temporarily screwed all of the cut pieces together and tested the validity of the mechanism. After several trials, we deemed the design to be adequate enough for our legged walker. During this week’s courses, we also decided to change our four legged walker into a six-legged walker. We decided to do so because we were worried about the stability of the robot when in motion. For instance, when a six legged walker is in motion, there is always three legs that are in contact with the ground, forming a stable triangular support for the robot to remain up straight. Furthermore, we also decided to simplify the leg design by employing only one for all six legs. This would alleviate our design and building process do give us more time to test the walker’s performance. Finally, we also thought about how the robot would turn. One of the Solution we came up with consisted of putting servos immediately above the feet of the walker robot. When the robot is in motion and three of its legs are lifted, we can then Use the servos to rotate the craft on the spot. In next week’s periods, we will continue to experiment with the walking mechanism and will further explore how to control the DC motors with a remote controller.
This week, we worked on the construction of the third prototype of our walker. Firstly, we decided use wood as the building material of this iteration. We also decided to implement the walking mechanism into this prototype. Furthermore, we also tried to attach the walking mechanism to a D.C. motor in order to test out the validity of the system. In addition to the motor, we also printed out mock-up legs that were meant to test the dimension of the legs and to prove the validity of the dimensions of our walking mechanism when attached to the legs. Finally, my partner also started on connected the motor to an Arduino in preparation for remote control of the motors. Unfortunately, the dimensions we started with were too short. Moreover, we realized we would have to build a horizontal platform for the motor to stay upstraight. Next week, we plan to complete this third prototype and move to building out final iteration.