julius.bernth

Julius Bernth
3rd Year Mechanical Engineering
I was born in Denmark but grew up in Washington DC, Berlin and, after while, Copenhagen. I came to the UK in 2005 and moved to London in 2008. I am doing a Mechanical Engineering degree at Queen Mary and I would like to work in the robotics industry. My 3rd Year Project supervisor is Dr. M. Hasan Shaheed. I am working with him on a master-slave robotic arm system.

First Parts Finished

 

Since my previous post, there have been three major developments in my project: I’ve finished and submitted all the blueprints for manufacturing in the university workshop; the first finished part was made; and I’ve started putting together the control software.

I submitted all the engineering drawings for the two robotic arms – both the master and the slave arms. They are going to be made in the university’s workshop. In fact, the handle used on the master arm has already been finished (pictured above). It was made using the university’s 3D printer – which I am still really amazed by. What surprised me the most was just how accurate the finished product was. Now I look forward to putting it to use in the finished prototype.

My supervisor, Dr. M. Hasan Shaheed, who was my personal tutor in the first and second year, and I have been looking at more applications for this technology. Dr. Shaheed has been looking into master slave technology for quite some time and so has had a lot of experience with it. So far, the healthcare industry seems the most promising and Dr. Shaheed has been working on contacting people in both the healthcare and robotics fields who may be interested in my project – all of which is extremely exciting.

As I also mentioned in my last post, I have begun working on the control system. My aim is to set up something simple and then implement it. Then the more complicated stuff can happen later. Like everything else in life, this process seems to be following the 80-20 rule: 80% of the effort goes into making it do 20% of what it’s supposed to – i.e. getting the motors to move. This is a pretty basic part of the control portion of my project, but so far, it seems to be the most time consuming. Because I am using new software and new equipment, there will always be a learning curve involved. The rest of the effort goes into fine tuning the controller so that it performs better. (I am sure that I will revise that statement in the next month once the controller starts to get more complex.)

First Data

The past couple of weeks have been occupied with crunching a lot of data and writing code. I’m now able to control one of my motors with my control program. I’ve included a basic graph showing the motor working. The details of the graph are not really that important – I know it’s meaningless without any scale information – but it’s just supposed to show that my set-up works (and short of a video, this is the next best thing). It may just look like a line on a bit of paper, but to me, this line represents the culmination four weeks of research, anxiety and general um-ing and ah-ing over how I should go about controlling the motors. So, when I saw this data for the first time, I was really, really happy.

It’s obviously not perfect. There is still loads of tuning to do, but I essentially have the foundation for my third year project. The rest of my project is going to be improving on this basic element and replicating it on the other motors once the rig is built – tweaking, essentially. There are going to be some challenges there, but I’ve dealt with that kind of stuff before when I made my initial prototype at home. Implementing the basic control system was always the biggest “if” in my mind, so doing the rest is not going to be nearly as intense.

The timing also couldn’t be better with Christmas coming up soon – I think I’ve earned a couple of days of lying on my parents’ couch and reading sci-fi books.

Robotic Arms: Starting a Challenging BEng Project

Robot snapshot

For my 3rd year BEng project, I am building a Master-Slave robotic arm that can be used for remote surgery/paramedical applications. Essentially, that means that I am building a robot arm (the slave arm) that a human being can directly control using a control arm (the master arm). The slave arm mirrors the movements of the master arm, which means it can perform complex motions without needing loads of programming and planning before hand – that is all done in an instant in the human brain. It’s kind of like that movie Real Steel, which aside from not being as bad as you might think (still not great, though), accurately summarises the argument behind master-slave technology.

The potential applications are pretty wide (aside from creating boxing robots, which would be awesome). Essentially, it just allows a human being to sort-of-directly interact with things they either aren’t strong enough to move or would rather avoid. That can mean anything from a battlefield, a nuclear reactor or outer space.

I got into this project because by the end of my 2nd year of doing mechanical engineering at Queen Mary, I had made up my mind that I wanted to get into control engineering. I’ve always been fascinated by how machines are able to react to their environment, either by direct command or through a sensor. Control engineering, specifically robotics, seemed to me to be the best way to work in that field and still be able to build stuff.

I spoke to my personal tutor, who is now my project supervisor, and he suggested looking into master-slave technology. I did and decided to build a simple rig at home using what I had learnt about servo motors in my second year instrumentation module (pictured above). I used an Arduino to read sensors on a master arm and relay signals to the servos in the slave arm so that their positions would match. Here is a video of it working: http://youtu.be/s1gio-IVdDc.

Now the challenge is to improve on the mechanical design and create a controller for it such that it can do something useful.  That means I need to put my own controller on it, which is the stage I am currently at in my project. The university have been really supportive of the project. They even extended the standard budget which I am extremely grateful for. Now it’s just a matter of hard work getting the thing up and running.

In my next post, I’ll give you some more details about how my project has been so far and where I am at the moment.

©2017 QMUL Student Blogs