Advances and Challenges in Robot Social Navigation
Advances and Challenges in Robot Social Navigation
ABSTRACT
Social navigation of service robots in human-populated environments is a key challenge to enable autonomous robotic platforms to support humans in their daily lives. Recent research on Machine Learning (ML) and Artificial Intelligence (AI) has increasingly focused on equipping robots with the ability to perceive, interpret, and adapt to complex social dynamics, enabling them to coexist safely with humans. Emerging methodologies leverage Large Vision Language models, Reinforcement Learning, and hybrid AI-model-based control systems to enhance robot navigation capabilities. For instance, multi-modal data processing enables robots to predict human trajectories, recognize social cues, and optimize navigation paths in real time. These approaches also prioritize safety, efficiency, and user comfort, aligning with the nuanced demands of real-world applications.
Despite these advances, significant challenges remain to achieve truly robust social navigation. One core challenge is defining and implementing the expected behaviors that balance navigation efficiency, friendliness, and safety, to ensure that robots navigate without causing discomfort. Moreover, standardizing metrics for evaluating social navigation performance remains a pressing issue, as traditional metrics such as time and distance fail to capture the complexity of human-robot interactions. Hence, there is a growing emphasis on developing socially-aware metrics that account for human preferences, proxemics, and specific subjects variations.
Moreover, current systems often struggle to reach scalability and adaptability over diverse environments with unstructured settings and unpredictable human behavior. Researchers are exploring ways to enable robots with higher-level semantic reasoning and predictive capabilities to address these limitations.
THE WORKSHOP
The workshop will take place on September 2 in conjunction with the European Conference on Mobile Robotics 2025 (https://ecmr2025.dei.unipd.it/).
This half-day workshop aims to bring together leading experts in mobile service robotics, social navigation, learning methods, and human-robot interaction to discuss the latest developments and challenges in the field.
The workshop aims to provide a platform for scientists, engineers, and practitioners to exchange ideas and share their research findings, with a focus on the practical applications of learning techniques in robotic social navigation.
It also looks at the latest innovations brought about by leading companies and startups in the field and serves as a bridge with the academic world.
CALL FOR PAPERS
TARGET PARTECIPANTS
Computer scientists and engineers
Data scientists and statisticians
Robotic engineers
Human-Robot Interaction and User Experience experts
Crowd and social models experts
TOPICS
Robot Social Behavior: challenges in defining the expected behaviors of robots in social contexts.
Evaluation Metrics: development of socially-aware metrics for performance evaluation.
Social robot navigation: novel solutions to navigate safely and respecting social norms in human crowded environments, from global planning to local controller methods. Novel approaches based on learning models or hybrid AI-model-based methods are welcomed.
Large Visual-Language Models (L-VLM) for social navigation: how to exploit and integrate L-VLM in a general perception and navigation framework for mobile robots.
Efficient perception: novel multi-modal data processing techniques to enhance the perception of mobile robots in social environments.
Scalability and adaptability in diverse and dynamic settings. Specific methods to address the generalization issues of learning models for social navigation are object of interest.
Real-world case studies and applications of autonomous mobile robotics with enhanced social behaviors in human-aware contexts.
Novel sensors and platforms for service robotics and indoor assistance.
GUIDELINES
We invite researchers and practitioners to submit extended abstracts of their work. The abstracts will be reviewed by the workshop organizers, and the authors of the selected papers will be invited to briefly present the work and participate in the poster session.
Authors will have the chance to engage with experts in the field of Robot Social Navigation, discuss about recent trends and challenges, and present their insights.
Demos associated with posters will also be encouraged by the workshop organizers to promote the participants' practical first-person involvement.
DEADLINES
Paper Submission Deadline: August 9th, 2025
Paper Author Notification: August 18th 2025
SUBMISSIONS
Papers: We welcome theoretical or empirical contributions describing ongoing projects or completed work. We encourage original ideas and contributions, also currently under review. The papers should be prepared with the IEEE RAS conference two-columns style template. The maximum length of papers is 4 pages excluding references.
Posters: Accepted papers will be presented with an oral pitch and a poster. Posters are required to follow A1 format.
Electronic submissions will be handled via EasyChair:
Call for papers: https://easychair.org/cfp/acrsn2025
Submission page: https://easychair.org/conferences/?conf=acrsn2025
If necessary, authors can directly contact the organizer and send ready manuscripts by the deadline in PDF format, email contacts are linked at the bottom of the page.
PROGRAM
9.00 - 9.10
09.10 – 09.50
-
09.40 – 10.30
-
10.30 – 11.00
11.00 – 12.10
12.10 – 12.50
-
12.50 – 13.00
Introduction
Keynote 1
Q & A
Company Presentations
Q & A
Coffee Break
Paper Pitch & Poster Session
Keynote 2
Q & A
Conclusive Speech
Workshop Organizers
Phani Teja Singamaneni
-
PAL Robotics
ALBA Robot
-
Selected Authors
Luis J. Manso
-
Workshop Organizers
-
Reasoning for better negotiations
in Socially Aware Robot Navigation
Social Robots at PAL Robotics
ALBA Autonomous Wheelchair : Project Empathy
-
-
Recent challenges and opportunities
in benchmarking social robot navigation
-
WORKSHOP COMMITTEE
Luis Merino is a Full Professor at the School of Engineering of the Universidad Pablo de Olavide (UPO), Seville, Spain, where he founded and leads the Service Robotics Laboratory. He holds a Ph.D. degree on Robotics, from the University of Seville under the supervision of Anibal Ollero. This thesis was awarded with the ABB Award to the Best Doctoral Dissertation on Robotics 2007 in Spain, given by the Spanish Committee of Automation (CEA, Robotics Group). His main research lines deal with cooperative robotic systems, including multi-robot systems and the cooperation between robots and persons. In these lines he has made contributions on new localization and navigation techniques, cooperative perception methods, decision making under uncertainties, and human-robot collaboration in social settings.
Fernando Caballero is a Full Professor at the School of Engineering of the Universidad Pablo de Olavide (UPO), Seville, Spain. His research career is focused localization systems and navigation in real environments using both aerial and grounded robots. In 2011, he founded, along with Professor Luis Merino, the Service Robotics Lab. During his career, he has participated in 9 European projects, 7 national projects, 7 regional projects and 8 project with companies, resulting in more than 30 journal publications and 20 high-impact international conferences.
Marcello Chiaberge is an Associate Professor of the Department of Electronics and Telecommunications at Politecnico di Torino (Turin, Italy). At the same university, he is also the Mechatronics Lab (LIM) Co-Director and Director and Principal Investigator of the Interdepartmental Center for Service Robotics (PIC4SeR). He has authored over 100 articles accepted in international conferences and journals and co-authored nine international patents. His research interests include the hardware implementation of neural networks and fuzzy systems and the design and implementation of reconfigurable real-time computing architectures.
Luis J. Manso is a Senior Lecturer (Associate Professor) in Computer Science at the College of Engineering and Physical Sciences of Aston University and a member of the Autonomous Robotics and Perception Laboratory from 2018. Before that, he was a Research assistant and Postdoctoral researcher at the University of Extremadura. He obtained his degree in Computer Engineering and his PhD from the University of Extremadura in 2009 and 2013, respectively. His research interests include geometric learning, active perception, human-robot interaction and navigation and sparse predictive world models.
Mauro Martini is a Postdoctoral Researcher with the Interdepartmental Center for Service Robotics (PIC4SeR) at Politecnico di Torino. He received a PhD cum Laude in Electrical, Electronics, and Communication Engineering at Politecnico di Torino with the thesis “Machine Learning for Perception and Autonomous Navigation of Service Mobile Robots”. His research interests currently involve learning methods for autonomous navigation in service robotics, focusing on perception and advanced behavioral policy learning. He previously organized and chaired workshops at international conferences (ECML-PKDD 2023).
Noé Pérez-Higueras is a postdoctoral researcher of the Service Robotics Lab at the Universidad Pablo de Olavide(UPO), Seville, Spain. He has participated in 4 FP7 research projects and 3 National projects. 20 publications as author or co-author can be counted, including 4 journal papers. His research lines are related to autonomous navigation of ground-robots, especially applied to areas shared with humans. He also employs Machine Learning methods to learn socially-compliant robot behaviors.
CONTACTS
Please contact us for any questions or doubts regarding the submission process and participation in the workshop.
Mauro Martini, Post-doc researcher, Politecnico di Torino: mauro.martini@polito.it
Noé Pérez-Higueras, Post-doc researcher, Pablo de Olavide University: noeperez@upo.es