Programming Robots for Activities of Everyday Life

Licentiate thesis, 2021

Text-based programming remains a challenge to novice programmers in all programming domains including robotics. The use of robots is gaining
considerable traction in several domains since robots are capable of assisting humans in repetitive and hazardous tasks. In the near future, robots will
be used in tasks of everyday life in homes, hotels, airports, museums, etc. However, robotic missions have been either predefined or programmed using
low-level APIs, making mission specification task-specific and error-prone. To harness the full potential of robots, it must be possible to define missions
for specific applications domains as needed. The specification of missions of robotic applications should be performed via easy-to-use, accessible ways, and at the same time, be accurate, and unambiguous. Simplicity and flexibility in programming such robots are important, since end-users come from diverse domains, not necessarily with suffcient programming knowledge.
The main objective of this licentiate thesis is to empirically understand the state-of-the-art in languages and tools used for specifying robot missions by
end-users. The findings will form the basis for interventions in developing future languages for end-user robot programming.
During the empirical study, DSLs for robot mission specification were analyzed through published literature, their websites, user manuals, sample
missions and using the languages to specify missions for supported robots.
After extracting data from 30 environments, 133 features were identified. A feature matrix mapping the features to the environments was developed
with a feature model for robotic mission specification DSLs.
Our results show that most end-user facing environments exist in the education domain for teaching novice programmers and STEM subjects. Most
of the visual languages are developed using Blockly and Scratch libraries. The end-user domain abstraction needs more work since most of the visual
environments abstract robotic and programming language concepts but not end-user concepts. In future works, it is important to focus on the development of reusable libraries for end-user concepts; and further, explore how end-user facing environments can be adapted for novice programmers to learn general programming skills and robot programming in low resource settings in developing countries, like Uganda.