Abstract
It is common knowledge among computer scientists and software engineers that ”building robotics systems is hard”: it includes applied and specialized knowledge from various scientific fields, such as mechanical, electrical and computer engineering, computer science and physics, among others. To expedite the development of robots, a significant number of robotics-oriented middleware solutions and frameworks exist that provide high-level functionality for the implementation of the in-robot software stack, such as ready-to-use algorithms and sensor/actuator drivers. While the aforementioned focus is on the implementation of the core functionalities and control layer of robots, these specialized tools still require extensive training, while not providing the envisaged freedom in design choices. In this paper, we discuss most of the robotics software development methodologies and frameworks, analyze the way robotics applications are built and propose a new resource-oriented architecture towards the rapid development of robot-agnostic applications. The contribution of our work is a methodology and a model-based middleware that can be used to provide remote robot-agnostic interfaces. Such interfaces may support robotics application development from citizen developers by reducing hand-coding and technical knowledge requirements. This way, non-robotics experts will be able to integrate and use robotics in a wide range of application domains, such as healthcare, home assistance, home automation and cyber–physical systems in general.
Highlights
Should one pose a question on the amount of skills needed to create, control or use a robot, the answer is quite straightforward: many! Robotics can be seen as an amalgam of mechanical engineering, electrical engineering and computer science, joining forces towards the construction and control of machines that interact with their environments
We introduce a series of abstractions applied for the robotics application development domain, as proposed by the model-driven engineering paradigm and the model-driven architecture (MDA) [38], so as to lower the complexity and required time to develop robotic applications
In the context of the proposed R4A architectural approach, robot resources are exposed in a homogeneous and standardized way, and automate the robot system and application design and development processes. We argue that this is expected to lead to a reduction in the development time and will allow for the wider penetration of low-cost robots in the market
Summary
Should one pose a question on the amount of skills needed to create, control or use a robot, the answer is quite straightforward: many! Robotics can be seen as an amalgam of mechanical engineering, electrical engineering and computer science, joining forces towards the construction and control (manually or autonomously) of machines that interact with their environments. Robotics can be seen as an amalgam of mechanical engineering, electrical engineering and computer science, joining forces towards the construction and control (manually or autonomously) of machines that interact with their environments. In this context, at least four skill sets are needed to create a robot from scratch. The second skill set includes electrical and electronic engineering knowledge for hardware design and implementation. This includes knowledge of various sensors, effectors and microcontrollers types, specifications and operation modes, ways to interconnect hardware devices (e.g., microcontrollers) with a variety of communication protocols, such as TTL, UART, SPI, and I2C [1], and electrical power supply and consumption basics. Software development capabilities involve knowledge of highlevel programming languages, such as C++, Python, Java and server-side JavaScript (e.g., NodeJs)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
