Research Team Proposes Human-Robot Co-Design Framework: ergoCub Adapts to Partners and Reduces Lower Back Strain

Researchers from Italy’s GenerativeBionics, the Italian Institute of Technology (IIT), and the University of Manchester have recently proposed a novel human-robot co-design framework that simultaneously optimizes both the physical structure and intelligent systems of humanoid robots to better align with human collaborative partners. The findings were published in the journal *Nature Machine Intelligence* and have already been applied to the new humanoid robot, ergoCub. Lead author Carlotta Sartore and corresponding author Daniele Pucci explained that in traditional robot design, hardware is treated as fixed, and control systems must compensate for hardware limitations—making the robot’s body a constraint on the development of intelligent behaviors, particularly evident in scenarios involving physical human-robot collaboration. Their new framework draws inspiration from nature, where body morphology and intelligence co-evolve, incorporating human users’ needs and characteristics directly into the design phase. The approach consists of two main stages, aiming to enhance safety, efficiency, and adaptability in human-robot collaboration. Experiments demonstrated that the ergoCub robot, developed using this framework, effectively reduces lower back load on humans during collaborative tasks.