Towards an interactive virtual reality human platform to create contact-rich full-body movements

Humans naturally perform many complex tasks. These include sitting down, picking up something from the table, and pushing a stroller. These operations involve different movements and require multiple contacts, which makes it difficult to program the robot to perform them.

Recently, Professor Eiichi Yoshida of Tokyo University of Science came up with the idea of ​​an interactive cyberphysical human (iCPH) platform to solve this problem. It can help understand and create human-like systems with contact-rich whole-body movements. His work has been published in Borders in Robotics and AI.

Professor Yoshida briefly describes the fundamentals of the platform. “As the name suggests, iCPH combines physical and cyber elements to capture human movement. While a humanoid robot acts as a physical twin to a human, a digital twin that exists as a simulated human or robot in cyberspace. The latter are modeled through techniques such as musculoskeletal and robotic analysis. The twins complement each other.”

This study raises several important questions. How can robots imitate humans? move? How can robots learn and simulate human behavior? And how can robots interact with humans in a smooth and natural way? Professor Yoshida addresses them within this framework.

First, within the iCPH framework, human motion is measured by quantifying the shape, structure, angle, velocity, and force associated with the movement of different body parts. In addition, human-made communication chains are also recorded. As a result, the framework allows for a general description of different motions through differential equations and creates a network of tangent motions on which a human figure can function.

Second, the digital twins learn this network through machine learning and model-based methods. They are connected to each other by the method of calculating the analytical slope. Continuous learning teaches the robot simulator how to perform contact sequences. Third, iCPH enriches the contact motion network through data enhancement and application of vector quantization technique. It helps to extract symbols representing the language of contact motion.

Thus, the platform allows to create contact motion in inexperienced situations. In other words, the robot can explore unknown environments and interact with humans using smooth movements that involve multiple contacts.

In fact, the author presents three challenges. These involve general descriptions, continuous learning, and contact motion symbolization. Navigating them is necessary to implement iCPH. Once developed, the new platform will have many applications.

“Data from iCPH will be made public and deployed to real-world problems to solve social and industrial problems. Humanoid robots can free people from many tasks that involve heavy burdens. and improve their safety, such as lifting heavy objects and working in hazardous environments.iCPH can also be used to monitor tasks performed by humans and help prevent related diseases. Finally, robots could be controlled remotely by humans via their digital twins, which would allow robots to perform large equipment installations and transport objects. ,” said Professor. .Yoshida, about the applications of iCPH.

Using iCPH as a starting point and with collaborative help from various research communities, including robotics, artificial intelligenceneuroscience and biomechanics, a future with humanoid robot not far.