Written by Victor Aladele
Senior, Electrical Engineering
Albert Dorman Honors College
This past summer (7th of June – 8th of August, 2015), I was selected to be a part of the MIT Summer Research Program (MSRP). I was assigned to the Computer Science and Artificial Intelligence Laboratory (CSAIL), where I worked directly under the supervision of a post doc, Dr. Robert MacCurdy; a faculty mentor, Professor Daniela Rus, who is the current director of CSAIL; and another student from Harvard, Youbin Kim. The goal of our project was to develop a 3D-printing procedure for the fabrication of hydraulic systems in small-scale robots.
Hydraulically-controlled soft actuators have recently become popular in small-scale robots due to their ability to more accurately imitate life-like motion, as well as their higher power-to-weight ratio. Unfortunately, most of these robots are limited in complexity and size due to constraints in conservative manufacturing techniques. In this project, we focused on testing the feasibility of the 3D-printing design approach to manufacture hydraulic systems. There are several limitations to this approach. These include how thin the components of the hydraulic system could be printed without causing fractures in the design parts and secondly, the challenge of designing two different parts very close together that will still be printed as two different components and not as a single part. Therefore, we printed several designs to see what the minimum measurements and tolerances were, to overcome these limitations. I designed external gear pumps using this fabrication technique. However, because I printed liquid within the gear pump, parts of the surfaces of the gears and the bottom surface of the gear housing got deformed due to the slushing of liquid over these surfaces. One way I was able to overcome this problem was to print thin support materials between the solid parts and the liquid. This prevented the liquid from being in direct contact with the solid parts. Having overcome this challenge, I was able to print a working 3D-printed gear pump without the need for any form of assembly. The success of this project could provide an attractive alternative to conventional methods, as this fabrication procedure eliminates the need for assembly of complex parts of a robot.
Although by the end of the nine weeks at MIT, we had not been able to make 3D-printed robots, I was able to manufacture a functioning gear pump to show the promise in this fabrication technique. At the end of the summer, I made a poster presentation at MIT, where I displayed the model I had made. My experience at MIT also involved attending breakfast and dinner sessions, where we (members of MSRP) listened to different professors at MIT who came and talked to us about their past and current research work. These professors also talked about their academic biographies and indicated some of the things that had helped them get to where they are now. Therefore, my experience at MIT was a holistic one and I am very confident that this has put me a huge step ahead in my preparation for graduate school.