Ship Motion Prediction: Smarter Robotic Fleet Navigation (PAL Robotics)
Challenge
PAL Robotics’ fleet of TIAGo robots previously operated with basic waypoint navigation and frequent manual interventions. Layout changes, reflective floors, and shared spaces with humans and forklifts often cause slowdowns, forcing the robots to rely on conservative speed limits for safety.
RoboSAPIENS Solution
Using the MAPLE-K architecture, PAL Robotics restructures its navigation logic into distinct adaptive layers. Sensor fusion (LiDAR, cameras, odometry) monitors localisation and map consistency; planning modules select optimal routes and speeds, while legitimisation enforces safety rules such as right-of-way priorities. All mission data and interventions are stored as knowledge for continuous improvement.
Impact
Testing and integration work are ongoing, but the new adaptive structure has already shown potential for smoother navigation and reduced manual overrides. It sets the stage for a fleet that can learn and optimise collaboratively.
Future Outlook
Next, PAL Robotics aims to extend the system to full pick–carry–place operations, powered by a smart fleet manager. Future versions will use large language models (LLMs) to translate plain-language commands into safe, executable missions, automatically validating safety rules and rerouting when necessary.
Social Sciences & Humanities Insights
RoboSAPIENS integrates Social Sciences and Humanities (SSH) research to ensure that adaptive robotics evolves in line with human values, ethics, and workplace realities. SSH activities explores how people interact with, trust, and accept robotic systems in each industrial context.
In the PAL Robotics Robot Navigation use case, investigations addresses safety perception and human comfort when robots operate alongside people in dynamic environments, emphasising clear communication of intent.
