ICNSC 2025

Special Session — Robotic Manipulators Beyond the Lab: Learning-Based Control, Perception, and Decision-Making

Scope and Objectives

With the recent surge in AI research and its impact on future of robotics and automation, robotic manipulators are undergoing a paradigm shift as they transition from controlled laboratory settings into the unpredictability of real-world environments. Traditional model-based control methods, while effective in structured spaces, often fall short in dynamic, unstructured, and partially observable environments. Meanwhile, advances in machine learning, reinforcement learning, and perception systems are equipping robots with the tools to adapt and operate more autonomously in such contexts.

This special session focuses on the intersection of control, AI, and perception with the aim of pushing robotic manipulation systems to operate with greater autonomy, resilience, and intelligence. We seek contributions that explore new learning paradigms, adaptive control frameworks, perception-action coupling, and systems that can function reliably beyond the confines of research labs. The goal is to highlight innovations that bridge the gap between theoretical development and real-world deployment, including applications in logistics, healthcare, disaster response, and manufacturing.

We encourage interdisciplinary approaches that combine insights from robotics, artificial intelligence, sensor fusion, control theory, and cognitive systems to address the multi-faceted challenges of autonomous manipulation in diverse and unpredictable environments.

Topics of Interest

Technical topics of this special session include but are not limited to:

Learning-based control strategies for robotic arms and mobile manipulators
Reinforcement learning and imitation learning for contact-rich manipulation
Perception-driven grasping and object interaction using deep vision models
Sensor fusion and tactile feedback integration in real-time control loops
Digital twins and simulation-to-reality transfer for manipulation policies
Safe, interpretable, and causal learning for robust robotic behavior
Edge computing and federated learning for distributed robotic manipulation
Adaptive and resilient control in uncertain and dynamic environments
Collaborative and multi-agent robotic manipulation
Applications in logistics, healthcare, inspection, and service robotics

Important Dates

→June 15, 2025: Special session paper submission due

August 15, 2025: Notification of acceptance

September 15, 2025: Camera-ready copy due

Submission Guidelines

Papers submitted to this special session should follow the standard formatting and submission guidelines of ICNSC 2025 which can be found on the Submission page: https://www.oulu.fi/icnsc2025/submission.html.

During the submission process, authors should select this special session by title (Robotic Manipulators Beyond the Lab: Learning-Based Control, Perception, and Decision-Making) to ensure their papers are directed appropriately.

Session Organizer

Session Organizer: Ameer Hamza Khan, Research Assistant Professor

Affiliation: Smart City Research Institute, Department of Land Surveying and Geo-informatics, The Hong Kong polytechnic University, Hong Kong.

Email: ameer-hamz.khan@polyu.edu.hk