In the context of automation, robots and AI accelerating globally, sensor technologies, soft materials and smart interactive systems are becoming the foundation for a new generation of robots in industry, services and healthcare. This is the outstanding message at the Seminar "Robot and Smart Automation" held on December 4 within the framework of VinFuture Science - Technology Week 2025.
Experts say that the human-shaped robot is getting very close to real life thanks to the advances of materials, AI and control. However, the biggest challenge lies not in technology but in society's level of trust in robots. According to Dr. Nguyen Trung Quan - University of Southern California, robots can only operate naturally if humans feel safe when standing next to them. The three important requirements he raised included physical safety, data security and social acceptance.
From a broader perspective, Gs. Do Ngoc Minh - University of Illinois Urbana-Champaign believes that automation can take away many jobs, so the acceptance of society depends on a sense of transparency and fairness when robots enter the labor market. This requires policies, education and communication in parallel with technology.
At the seminar, soft materials were considered the focus. Gs. Kurt Kremer - Polymer Max Planck Institute commented that soft polymeas bring flexibility, ease of manufacture and the ability to react to the environment, allowing for the creation of soft drivers to help robots operate more sophisticatedly.
When polyme is designed into a gel or brush structure, the material can stretch, change shape and take on complex mechanical tasks. Some materials are also electrical or have high electrolytes, paving the way for the development of organic electronic devices and robots simulating the nervous system, creating a foundation for a safe and cost-effective generation of robots.
Gs Ho Young Kim - Seoul National University believes that the biggest challenge for robots today is handling soft materials such as clothes, food or medical supplies. Just a small change in the clip point can also change the shape of an object, creating countless parameters that AI has difficulty controlling.
His research team is developing elastic film removal technology for robots to stably process layers of fabric and soft objects. According to him, robots need to overcome four barriers: accurate awareness of material state, sophisticated manual transmission, flexible control system and expansion capability during production.
In the direction of developing human-shaped robots, Gs Tan Yap Peng - Principal of VinUni commented that robots could become a popular force by 2050 when the world can reach the one billion robot mark. However, robots today are mainly programmed for individual tasks.
To move towards multi-purpose robots, technology must learn from large language models, use a huge video data warehouse to build the ability to understand the physical world. The biggest limitation is that robots do not have the same memory as humans. His team proposed a memory section model for robots to retrieve experiences when encountering new tasks, helping to reduce errors and increase the ability to complete long tasks.
Along with AI, the energy requirement, hand skill, and the ability to self-dygnosis and operate safely are considered important problems for the next 30-50 years. At the implementation level, Dr. Nguyen Trung Quan believes that a human-shaped robot that wants to operate effectively must have good hardware, strong AI and a safe implementation system. According to him, the human-shaped and physical intelligence robot market could reach 10,000 billion USD in the next 10 years, especially when many countries are lacking labor.
According to experts, Vietnam has a great advantage when entering the field of human-shaped robots thanks to the young force of engineers and the rapid increase in demand for robot applications in industry and services. Early participation in research and development of a robotic platform is considered an important step to seize new opportunities for physical intelligence.
