Profile
Kotaro Noji
Spatial Robotics Research Center
Graduate School of Science and Technology
Joined Fujitsu in 2024
My Purpose: Conducting cutting-edge research with innate curiosity, an open mind and a readiness to challenge conventional ideas!
Article|2025-12-23
My pathway to the future, interwoven with the concepts of "Growth" and "Enjoyment"
When I was a child, I vaguely thought, "I want to be an amazing person." During my soccer training, which lasted from elementary school through high school, I was strongly drawn to the highly skilled players, who played with apparent enjoyment and were respected by those around them. At the time, I simply thought that anyone who was good at soccer was amazing, but looking back now, I believe it was their joyful attitude that attracted me. Out of this experience grew the idea that improvement through continuous effort leads to even greater enjoyment – something that has become crucial throughout my career.
My interests weren't limited to soccer; I was also keen on programming. Originally, I aspired to be an engineer, but my encounters with researchers through collaborative research in graduate school significantly changed my career path. I developed a strong admiration for them, because they possessed "their own strengths" in the form of intense specialization, and seemed genuinely to enjoy facing daily challenges by making the most of their skills. On top of this, n research, one can approach things from multiple standpoints – including implementation skills as well as an academic standpoint, creating an environment where solving more fundamental problems can be truly enjoyed. For these reasons, I really wanted to achieve a positive cycle of "growth" and "enjoyment" in the world of R&D.
Aiming for a world where humans and robots collaborate and coexist
I am currently a member of the Spatial Robotics Research Center, which was established in April 2025. Our center aims to realize a world where robots can behave intelligently without posing a danger to humans, by comprehensively recognizing their entire spatial environment using deployed sensors. Within this context, I am researching "maps" that enable robots to act intelligently in human living spaces. Specifically, this involves developing technology to reflect the behavioral history of "when, where, and what a person did" onto a map.
When robots operate in real-world environments, correctly understanding their surroundings is a significant challenge. For example, if instructed to "get the cup from the kitchen," a robot will try to complete the task in the shortest time and by the shortest route. As a result, if a child happens to be on that route, the robot might prioritize task completion and potentially collide with the child. By applying this technology, information about the times and locations where children are frequently present is recorded on the map as behavioral history. This allows the robot to determine, "It's better not to take this route into the kitchen at this time." Consequently, instead of the shortest route, the robot can choose a safety-conscious route to retrieve the cup. In the future, my goal is to realize a convenient society where humans and robots smoothly collaborate, for instance, in areas like logistics and elderly care where labor shortages are a challenge, enabling robots to safely, securely, and efficiently support people.
This photo is a demonstration of our "Spatial World Model". It integrates cameras placed in an environment, with cameras mounted on robots – enabling us to imagine the future of the entire space, predicting a person's actions and intentions to enter a restricted area. Unlike conventional technology that detects events after they occur, its key feature is the ability preemptively to coordinate multiple robots based on the predicted future. For more details, please refer to the following content (*1)(*2)(*3).
Constantly engaging with cutting-edge technology, aiming to be a leader in point cloud processing
In R&D, I have two major personal goals.
The first is to become a leading expert in the field of point cloud processing. Point cloud processing is a technology that digitizes, analyzes, and recognizes the shape of entire environments, including 3D space and various objects within it, as a high-density set of points (point cloud data). I am currently working on developing technology to estimate human posture from point cloud data collected by robots, and recognize human behavior from the sequence of these postures. This can be utilized for recognizing the behavioral history of when, where, and what a person did. Point cloud data is generated by irradiating objects with laser light and measuring the reflected light to determine the distance and shape of the object. Therefore, compared to camera images, point cloud data is more mindful of individual privacy. In addition, it allows for highly accurate measurements even in dark conditions, making it a technology with significant potential for monitoring in privacy-sensitive locations such as hospital rooms and care facilities. I aim to develop this technology further and contribute to realizing a society where robots that support human safety are commonplace in every space where people are active.
My second goal is to keep up with cutting-edge technology continuously and become a researcher, who possesses both capability and peace of mind. A senior colleague, who fostered my growth, serves as my future role model. This senior colleague achieved a high level of balance in research skills, communication ability, and personal life, always maintaining a calm demeanor while supporting my inexperienced self. The desire to absorb as much as possible every day, to become a little more like my senior colleague all the time, is currently my driving force. In fact, before joining the company, I had no experience with actual robot hardware. So, I decided that the only way to learn was to get actively involved, and I proactively participated in external lectures and development seminars to acquire knowledge and skills. As a result, in the final project of a Physical AI Intelligent Robotics Course, I was proud to be recognized with an excellence award for demonstrating creativity and technical prowess. In addition to the robot development techniques learned in the course, I applied MCP (Model Context Protocol) to control a robot with natural language commands. This experience also provided me with the opportunity to join an invitation-only community, allowing me to experience significant growth and new worlds in a short period.
My final project of the Physical AI Intelligent Robotics Course
Generative AI: my current favorite hobby
My hobbies are diverse, including soccer, basketball, camping, programming, gadgets, anime, and manga, but what my current passion is generative AI. By utilizing generative AI, I can create another me – or rather, a more capable supporter than my current self – which dramatically accelerates brainstorming and information gathering. As a result, it feels like I've gained far more time than if I were working alone. The fact that I, a beginner in robot development, was able to advance my development is, I believe, because I constantly pursued new technological trends and utilized the knowledge and skills I gained en route. While I'm currently absorbed in generative AI, even if the next technological trend emerges in the future, I always want to chase the cutting edge and enjoy the very act of acquiring new technologies. In this way, by learning new technologies and leveraging them for growth, I can undertake even more enjoyable and stimulating challenges. I intend to continue accelerating this positive cycle of "growth" and "enjoyment."
I took this photo during an osteopathic session with an osteopathic practitioner (*4), who is well-versed in body structure and AI technology. He provided highly valuable insights and advice from his perspective regarding body movement design in Physical AI.
Messages from colleagues
Kotaro possesses an exceptionally sharp antenna for new technologies, and he consistently brings vitality and fresh perspectives to our team. His excellent information-gathering abilities are a major asset in the rapidly changing field of AI. Furthermore, his sincere dedication to even the most painstaking tasks during on-site robot testing earns the deep trust and reassurance of the team. Moving forward, we anticipate that he will combine his practical capabilities and inquisitive spirit with deep insights into the future, allowing him to soar in the field of Physical AI. (Jiang Shan, Research Director, Spatial Robotics Research Center)
Kotaro is a young researcher with a strong sense of curiosity. He is incredibly quick at catching up with cutting-edge technologies and always shares his findings with enthusiasm, providing us with learning and inspiration. He is also highly motivated in the internal study sessions we participate in together, and thanks to Kotaro, there have been many times when my understanding of the study session content has deepened. As a colleague and a respected researcher, I genuinely look forward to his future successes. (Yuta Watanabe, Data & Security Research Laboratory)
Titles, numerical values, and proper nouns in this document are those reported when this interview was made.