Advanced Competitive Science (ACS)
At Benilde- St. Margaret's High School, there is a program called Advanced Competitive Science or ACS that I have taken since my sophomore year. Essentially, it is a introduction to engineering class with a hands on approach. The pedagogy is what distinguishes this program from others. We are provided with a curriculum that we must use to learn on our own. It has been proven through research that the most effective method of learning is through self discovery, which is what they're after in this program.
These are the course descriptions if you are interested. (source-http://www.bsmschool.org/academics/departments/acs/acs-courses/)
Engineering I
In this first-year course, students develop an understanding of effective problem solving and machine design while exploring fundamental engineering concepts including statics (objects in equilibrium), dynamics (objects in acceleration), sensor and control system function, and elementary embedded logic programming.
Engineering II
This second-year course expands on the skills developed in Engineering I and introduces students to designing with 3D CAD software, fabrication with rapid prototyping, and preliminary development of an advanced project that will carry over to Engineering III.
Engineering III
In this third-year course students will continue to expand skills and knowledge in engineering design and problem solving as well as begin exploration into higher levels of embedded logic design and programming. Students will complete a high-level project (begun in Engineering II) with the possibility of qualifying for the ACS Engineering Travel Team.
This whole program revolves around the field of search and rescue robotics, which luckily for me is my field of study. At the beginning of each year, we receive a kit of various components (mainly Lego) including motors, sensors, etc. Using the engineering design process, we build a robot that can navigate a representation of a disaster area. The whole structural aspect of the robot is built from Legos and then we use off the shelf sensors and motors (continuous rotation servos). We also have the luxury of being able to print custom parts made of ABS plastic. In order to do this, we learned how to use the CAD program, Solidworks, and a laser cutter. For control/ processing, we use a custom "stack" that is comprised of a SRV-RCM board, a Blackfin camera module, and Maxtronix Communication Board. To see if our mechanical design will work, the robot is initially teleoperated. Once we are able to successfully navigate through the search and rescue area, we make it autonomous using C language derivative, PicoC.
So there is an overview of the program. Now onto our work.
These are the course descriptions if you are interested. (source-http://www.bsmschool.org/academics/departments/acs/acs-courses/)
Engineering I
In this first-year course, students develop an understanding of effective problem solving and machine design while exploring fundamental engineering concepts including statics (objects in equilibrium), dynamics (objects in acceleration), sensor and control system function, and elementary embedded logic programming.
Engineering II
This second-year course expands on the skills developed in Engineering I and introduces students to designing with 3D CAD software, fabrication with rapid prototyping, and preliminary development of an advanced project that will carry over to Engineering III.
Engineering III
In this third-year course students will continue to expand skills and knowledge in engineering design and problem solving as well as begin exploration into higher levels of embedded logic design and programming. Students will complete a high-level project (begun in Engineering II) with the possibility of qualifying for the ACS Engineering Travel Team.
This whole program revolves around the field of search and rescue robotics, which luckily for me is my field of study. At the beginning of each year, we receive a kit of various components (mainly Lego) including motors, sensors, etc. Using the engineering design process, we build a robot that can navigate a representation of a disaster area. The whole structural aspect of the robot is built from Legos and then we use off the shelf sensors and motors (continuous rotation servos). We also have the luxury of being able to print custom parts made of ABS plastic. In order to do this, we learned how to use the CAD program, Solidworks, and a laser cutter. For control/ processing, we use a custom "stack" that is comprised of a SRV-RCM board, a Blackfin camera module, and Maxtronix Communication Board. To see if our mechanical design will work, the robot is initially teleoperated. Once we are able to successfully navigate through the search and rescue area, we make it autonomous using C language derivative, PicoC.
So there is an overview of the program. Now onto our work.
Initially, my partner, Ben Lenington, and I decided we wanted to stray from the norm of using treads; the reason being that the small radius of drive wheel for the tread was not large enough to climb over the block field. So we used the large balloon tires for its large radii in combination with suspended to skids to reduce friction, but enable flexibility . To compensate for the lack of surface area of the balloon tires, we used two on each side making it a dually.
Constructing it was straightforward. Everything snapped together according to our design. We built the chassis in approximately one week, got our SRV stack and servos configured and the antenna mounted. We were ready to test. Driving on flat ground was very smooth. Unfortunately, the rest of the results of the tests were less than acceptable. Due to the lack of traction and surface area of the balloon tires in combination with the heavy weight, the robot struggled to climb the block field, stairs, and thirty degree ramp. After initial testing, we made a few tweaks to the design. In attempt to increase the traction, we used the treads, which you can see in the picture of Iteration 1.2. It did improve the robot's climbing capabilities, but it created several other problems. The treads presented numerous inefficiencies and made the robot very difficult to control. It would have been impossible to autonomize. Ben referred to it as "a bowl in a China shop". It was a very solid and robust design, just too large and heavy for this application. After further consideration, we decided the design was beyond salvageable and commenced designing iteration 1.3 from scratch. |
|
We did not consider the first iteration to be a failure though. We learned many things from it and what we must to do to conquer the course with ease.
- It must be lightweight in order to climb the ramp and other obstacles - The drive components must have a lot of surface area to climb the obstacles as well as the ramp - The treads must have a positive slope in the front and back in order to climb over the block field in both directions - The center of mass should be towards the front in order to prevent it from tipping backwards - There must be enough ground clearance to prevent it from catching on blocks. - A higher gear ratio can be used to increase speed (more than enough torque) - The majority of the mass should be as low to the ground as possible to prevent it from tipping backwards So with all these new design requirements in mind, we constructed a brand new robot in just under two weeks. It's simple, elegant, and it works remarkably |
|