research
The things we work on in the @Home team should be scientifically sound and relevant. The methods we implement for solving the problems in the @Home scenarios need to reflect the current cutting edge of the respective fields; in other words, our work should not consist of simple hacks that may work in some cases, but may not work in general.
How to proceed when you are given a task to work on
The only way of doing proper scientific work is to become familiar with the existing work in your field of work before you even start thinking about robot-specific implementations; in other words, coding should certainly not be the first thing you are concerned with when you are given a task to work on. As a way of ensuring the scientific relevance of your work in the context of the @Home team, you will thus need to follow a few rules:
When you are given a task, if you know very little about the topic, start identifying the important and most recent literature on it.
Once you have identified a sufficient amount of related work (the meaning of sufficient depends on the topic - some research areas are more active than others - so you will have to use your own judgment and the help of the scientific moderators regarding this matter), you will need to perform a comparative analysis of the existing approaches; this will help you decide which methods are the most promising ones for being applied and extended to the @Home scenarios. The comparative analysis you perform should ideally be quantitative; in principle, you should be able to use already existing method implementations while performing the analysis, but you may need to implement some of them yourself if you think they seem to be really promising. In some cases, qualitative analysis could be sufficient as well, but you should consult the scientific moderators about which type of analysis is more appropriate in a given case.
After finishing your comparative analysis, talk about it with the scientific moderators; if they are satisfied with it, you can proceed with your robot-specific implementation. This means that you can start implementing things on the robot only after you have analysed the related work extensively.
Once you are done with your implementation, please have a look at the pull request diagram in order to make sure that your code gets accepted in the stable branch of the common repository.
Scientific moderators
Scientific moderation is currently performed by Alex Mitrevski (C201) together with Prof. Paul G. Plöger (C224); your comparative analysis will thus be reviewed and approved by them.
Scientific reproducibility and publications
The fact that we are an applied science institute gives us the unique opportunity to test and evaluate the more conventional research produced at other institutes in addition to the research we are doing ourselves. Given that reproducibility is one of the main pillars of science, the @Home team has the chance to do one of the most important aspects of the scientific process - verifying that published research is what it claims to be. In addition to ensuring that our robot utilises the cutting edge methods, doing your homework properly could thus also lead to many scientific publications from our group.
Last updated