Open-source software for learning-based robot control: from simulation to real-world deployment
Angela Schoellig
Technical University of Munich (TUM)
8th International Workshop on Robotics Software Engineering (RoSE’26)
- co-located with ICRA 2026
- in-person event in Vienna, Austria
- June 1st, 2026 (Monday)
Theme & Goals
Robotics is one of the most challenging domains for software engineering. Deploying even simple applications requires integrating solutions from several domain experts, including navigation, path planning, manipulation, localization, human-robot interaction, etc. Integration of modules provided by the respective domain experts is one of the key challenges in engineering software-centric systems, yet only one of the cross-cutting software concerns that are crucial to robotics. As robots often operate in dynamic, partially observable, and populated environments, additional challenges include adaptability, robustness, safety, and security.
RoSE aims to bring together researchers and practitioners to identify new frontiers in software engineering for robotics, discuss challenges arising from real-world applications, and transfer the latest insights from research to industry.
Topics of Interest
RoSE 2026 seeks contributions addressing, but not limited to, the following topics related to robotics software engineering:
- AI agents in the context of robotic software systems
- AI-enhanced development of robotics systems
- Analysis of challenges in robotic software engineering
- Best practices in engineering robotic software
- Continuous integration and deployment in robotics
- Domain-specific languages and tools for the development of robotic software
- Empirical studies of robotics software and software tools, including its usability
- Engineering (heterogeneous) multi-robot systems.
- Identification and analysis of design principles promoting quality of service (e.g., performance, energy efficiency)
- Lessons learned in the engineering and deployment of large-scale, real-world integrated robot software
- Machine learning applied to robotic software
- Metrics measuring non-functional properties (e.g., robustness, availability, etc.) and their application to robotic software
- Mining software repositories of robotic systems
- Model-Driven Engineering methods and techniques applied to robotic software
- Processes and tools supporting the engineering and development of robotic systems
- Software architecture of robotic systems
- State-of-the-art research projects, innovative ideas, and field-based studies in robotic software engineering
- Static and dynamic analysis of robotic software
- Usability studies of robotics software and tools
- Variability, modularity, and reusability in robotic software
- Validation and verification of robotic software
Invited Speakers
more to be announced soon…
Important Dates
- March 8th, 2026: Workshop Papers Submission
- April 15th, 2026: Workshop Papers Acceptance Notification
- April 30th, 2026: Workshop Papers Camera Ready
- June 1st, 2026: Day of the Workshop
Submission Guidelines
Prospective participants are invited to submit:
-
Research Idea papers (structured one-page abstract) - we invite early‑career researchers to submit structured one‑page abstracts for presentation at our workshop. Submissions will not undergo peer review, but their quality and relevance will be closely evaluated by the organizing committee, who will invite the shortlisted ones to be shared within the event to foster open discussion of new ideas and forthcoming paper plans, remaining unpublished. Format: the one-page abstract must follow the ICRA double-column IEEE style (i.e., the standard ICRA manuscript format). (ras.papercept.net)
-
Industry reports (one-page abstract) - we invite industry contributors to submit one‑page abstracts that showcase fresh ideas, emerging trends, or real‑world experiences from the corporate sector. Submissions will not undergo peer review, but their quality and relevance will be closely evaluated by the organizing committee, who will invite the shortlisted ones to be shared within the event. For the shortlisted reports, publication on the RoSE website is optional, allowing authors to keep their work internal or make it publicly accessible. Format: the one-page abstract must follow the ICRA double-column IEEE style (i.e., the standard ICRA manuscript format). (ras.papercept.net)
-
Work-in-progress papers (six-page papers) - we invite submissions of work‑in‑progress papers, limited to six pages (including references), that present preliminary results, open challenges, lessons learned, or descriptions of tools and projects. All papers will undergo peer review, and accepted contributions will be published on the RoSE website. The accepted papers will be submitted for inclusion in a proceedings volume. Format: the six-page papers must follow the CEUR-WS (CEURART) style. (CEUR-WS (CEURART) template)
The best paper authors will be invited for submission of extended results in the Frontiers research topic linked to the workshop and will be awarded a voucher for free publication in case of acceptance.
Templates and submission process
-
One-page abstracts (Research Ideas + Industry Reports): must use the ICRA submission format, as described in the PaperCept manuscript preparation guidelines. (ras.papercept.net)
-
Work-in-progress papers: must follow the CEUR-WS (CEURART) template and must consider a single blind submission process. All work-in-progress papers will be reviewed by the program committee on the basis of technical quality, relevance, significance, and clarity.
-
All submissions should be uploaded electronically in PDF format through EasyChair at this link: TBD
On the use of AI (Artificial Intelligence) or AI-assisted technologies:
When submitting to RoSE 2026, the authors acknowledge that they comply with the Generative AI usage policy, based on existing policies proposed by IEEE, ACM, and Springer. It is forbidden to:
- List Generative AI tools and technologies, such as ChatGPT, as authors of works;
- Use texts or sections entirely produced by generative AI tools
It is allowed (with explicit disclosure in the acknowledgments) to:
- Use generative AI tools to create parts of the content, with disclosure in the paper acknowledgments indicating what was generated and which tool was used. It is important to check the terms of use of the tool, which is the responsibility of the authors. For example, in the acknowledgments, you can use: ChatGPT was used to generate the first paragraph of Section 3 and to generate Table 3.2.
It is allowed (no need to disclosure):
- Use AI or AI-assisted technologies to improve the quality of images in terms of contrast and clarity;
- Utilize generative AI tools to edit and improve the quality of your existing text (similar to an assistant like Grammarly to improve spelling, grammar, punctuation, clarity, and engagement).
Organizing Committee
- Michel Albonico (Technological Federal University of Paraná, Brazil / University of Southern Denmark (SDU), Denmark)
- Hannan Ejaz Keen (XITASO GmbH, Germany)
- Ricardo D. Caldas (Gran Sasso Science Institute (GSSI), Italy)
Program Committee
TBD
