Humans naturally perform many complex tasks. This involves sitting down, picking up an object from a table, and pushing a cart. These activities involve different movements and require multiple contacts, making it difficult to program robots to perform them.
Recently, Professor Iiichi Yoshida of the Tokyo University of Science put forward the idea of an electronic physical human interaction (iCPH) platform to address this problem. It can help understand and generate human-like systems with rich full-body contact movements. His work has been published in Frontiers in robotics and artificial intelligence.
Professor Yoshida briefly describes the basics of the platform. “As the name suggests, iCPH combines physical elements with electronic elements to capture human motions. Whereas a robot acts as a physical twin of a human, a digital twin exists as a simulated human or robot in cyberspace. The latter is modeled through techniques such as skeletal and robotic muscle analysis. The twin complements each other.”
This research raises several key questions. How can a person imitate human movement? How can robots learn and mimic human behaviour? How can robots interact with humans smoothly and naturally? Professor Yoshida addresses them in this context. First, within the iCPH framework, human motion is measured by determining the shape, structure, angle, speed, and force associated with the motion of different body parts. In addition, the sequence of communication made by the human being is also recorded. As a result, the framework allows for the general description of different motions through differential equations and the generation of a contact motion network on which a human can act.
Second, the digital twin learns this network through both model and machine learning approaches. They are linked together by the method of calculating an analytical gradient. Continuous learning teaches the robot simulator how to perform the communication sequence. Third, iCPH enriches the communication traffic network via data augmentation and applies vector quantization technology. It helps in extracting the symbols that express the language of the communication movement. Thus, the platform allows the generation of contact traffic in inexperienced situations. In other words, robots can explore unknown environments and interact with humans using smooth motions that involve many contacts.
In fact, the author poses three challenges. These relate to generic descriptors, continuous learning, and call traffic code. Its navigation is essential to achieving iCPH. Once the new platform is developed, it will have many applications.
“Data from iCPH will be announced and disseminated in real-life problems to solve social and industrial issues. Humanoid robots can free humans from many burdensome tasks 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 work-related diseases.Finally, humans can be remotely controlled by humans through their digital twins, which will allow humans to install large equipment and move objects,” says Professor Yoshida about iCPH applications.
With iCPH as ground zero and with the help of collaborations from different research communities, including robotics, artificial intelligence, neuroscience, and biomechanics, the future of humanoid robots is not far away.