[Paper Review] Towards An Architecture-Centric Approach to Manage Variability of Cloud Robotics

Zhang, Lei, Huaxi Zhang+5 more|arXiv (Cornell University)|Jan 13, 2017

Robotics and Automated Systems3 citations

TL;DR

This paper proposes CRALA, a domain-specific architecture description language for cloud robotics that enables architecture-centric design and variability management across specification, configuration, and assembly levels. By integrating Cloud deployment constraints (e.g., scheduling, networking) into a structured modeling framework, CRALA supports automated generation of deployable robotic system architectures with enhanced reusability and configurability through Ecore and Sirius-based tooling.

ABSTRACT

Cloud robotics is a field of robotics that attempts to invoke Cloud technologies such as Cloud computing, Cloud storage, and other Internet technologies centered around the benefits of converged infrastructure and shared services for robotics. In a few short years, Cloud robotics as a newly emerged field has already received much research and industrial attention. The use of the Cloud for robotics and automation brings some potential benefits largely ameliorating the performance of robotic systems. However, there are also some challenges. First of all, from the viewpoint of architecture, how to model and describe the architectures of Cloud robotic systems? How to manage the variability of Cloud robotic systems? How to maximize the reuse of their architectures? In this paper, we present an architecture approach to easily design and understand Cloud robotic systems and manage their variability.

Motivation & Objective

• To address the lack of systematic architectural modeling for cloud robotics systems with high variability.
• To enable reuse and configuration of robotic system architectures across different Cloud deployment environments.
• To integrate Cloud-specific non-functional requirements (e.g., scheduling, networking) into architectural descriptions.
• To support automated generation of deployable architectures from high-level models.
• To provide a structured, extensible framework for managing architectural variability in cloud robotics.

Proposed method

• Designing a three-level architecture model: specification (component roles), configuration (component/services), and assembly (instances) to manage variability.
• Defining a metamodel for CRALA using Ecore to represent architectural elements and their relationships.
• Integrating Cloud deployment concerns—such as scheduling, networking (e.g., OpenStack Neutron), and physical machine placement—into architectural models.
• Using Sirius to generate graphical modeling workbenches for visual editing and automatic model-to-diagram rendering.
• Implementing the CRALA tool suite on Eclipse-based EMF and Sirius platforms for model creation and visualization.
• Automatically generating example architectures (e.g., Arch1) from defined relationships and deployment constraints.

Experimental results

Research questions

• RQ1 How can cloud robotics systems be systematically architected to manage variability across different deployment configurations?
• RQ2 How can Cloud-specific non-functional properties (e.g., scheduling, network topology) be formally integrated into architectural descriptions?
• RQ3 How can architectural variability be managed across multiple abstraction levels (specification, configuration, assembly)?
• RQ4 Can a domain-specific language (CRALA) enable automated generation and deployment of cloud robotic system architectures?
• RQ5 How can component reuse and configuration be supported through a structured modeling approach in cloud robotics?

Key findings

• CRALA successfully models cloud robotics architectures across three abstraction levels: specification, configuration, and assembly, enabling structured variability management.
• The integration of Cloud deployment constraints—such as VM placement and network topology—into architectural models improves system control and clarity.
• The CRALA tool suite, built on EMF and Sirius, enables automatic generation of graphical models from metamodels, supporting visual design and validation.
• Example architectures (e.g., Arch1) were automatically generated based on defined relationships and deployment rules, demonstrating feasibility and reusability.
• The approach supports horizontal variability across architecture levels and vertical variability through component realizations, enhancing configurability.
• The implementation shows that architectural descriptions can be linked directly to Cloud deployment aspects, enabling better control and automation of robotic system deployment.