We are proud to share the invited article “Robot Contests as a Catalyst for Robotics Science”, published in the Annual Review of Control, Robotics, and Autonomous Systems (2025). Authored by Pedro U. Lima, Carlos Azevedo, and Rodrigo Serra, this work reflects on the crucial role of robot contests in driving scientific progress and innovation in the field.

While traditional robotics research often emphasizes subsystem optimization or theoretical design, robot contests provide a unique platform for benchmarking complete robot systems. By encouraging integrated system development and tackling realistic challenges, competitions push the boundaries of what robots can achieve. The article discusses how these contests, when designed following best practices, stimulate progress across domains, strengthen education by promoting teamwork and problem-solving, and support technology transfer — with several notable spin-off companies originating from competition environments.

The paper surveys well-established robotics contests as sources of diverse approaches to performance assessment and best practices, while also introducing alternative formats such as cooperative competitions. In particular, it highlights the potential of these approaches to foster transferability across robotic platforms and tasks, supported by benchmarking metrics. A dedicated section also reflects on the role of the euROBIN Coopetitions as an innovative contribution to this evolving landscape.

Learn more about the scientific paper here: https://www.annualreviews.org/content/journals/10.1146/annurev-control-022624-012944

Congratulations to the authors Pedro U. Lima, Carlos Azevedo, and Rodrigo Serra for this insightful contribution that underscores the importance of contests as catalysts for robotics research and innovation.

We are pleased to announce that the article “Perspective-Shifted Neuro-Symbolic World Models: A Framework for Socially-Aware Robot Navigation” has been accepted at the 2025 IEEE International Conference on Robot & Human Interactive Communication (RO-MAN). This work was carried out under euROBIN’s WP7, led by PhD student Kevin Alcedo and co-supervised by Pedro U. Lima and Rachid Alami.

Robots navigating alongside humans must not only reason under uncertainty but also account for the beliefs, intentions, and perspectives of those around them. Traditional egocentric navigation can be formulated as a Markov Decision Process (MDP), yet social navigation requires extending this to a Partially Observable MDP (POMDP), where agents lack direct access to others' mental states. To address this challenge, the paper proposes a neuro-symbolic, model-based reinforcement learning architecture that enables effective belief tracking in partially observable environments. It further introduces a perspective-shift operator for belief estimation, while also leveraging recent advances in Influence-based Abstractions (IBA) to improve reasoning in structured multi-agent settings.

This contribution opens new directions for socially-aware navigation, bridging symbolic reasoning and neural learning methods in robotics.

More information about the article: https://arxiv.org/abs/2503.20425

Congratulations to the authors Kevin Alcedo, Pedro U. Lima, and Rachid Alami for their innovative work!

We are excited to announce the acceptance of the article “Large Language Model-Based Robot Task Planning from Voice Command Transcriptions” at the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2025). This work was developed within euROBIN’s research activities on service robotics (WP2), focusing on enabling robots to perform domestic tasks from natural language instructions.

One of the central challenges in building a General Purpose Service Robot (GPSR) lies in interpreting human commands, which often contain speech recognition errors and incomplete information. This paper presents an end-to-end pipeline that leverages a Large Language Model (LLM) to directly translate instruction transcripts into coherent and context-aware action plans. By integrating environmental information into the input, the system is able to generate more efficient task executions.

Key highlights of the paper include:

• An LLM-based approach to bridge natural language instructions and robot task planning

• Integration of environmental context for improved efficiency

• Validation through a simulator based on Generalized Stochastic Petri Nets, with a success rate of ~55% on the ALFRED dataset, including unseen environments

• Successful real-world deployment at RoboCup 2024 in Eindhoven, achieving 2nd place in the GPSR task

You can access the full paper here: https://afonsocerto.com/files/iros2025_llm_robot_planning.pdf

Congratulations to the authors Afonso Certo, Bruno Martins, Carlos Azevedo, and Pedro U. Lima for this significant contribution to advancing natural language understanding in robotics!