Service robots are developed to assist in both home and commercial settings. These robots collaborate with humans to accomplish a variety of tasks, and their design emphasizes safety and adaptability. To effectively perform their roles, they require capabilities such as voice and image recognition, manipulation skills, human-robot interaction, and flexible task-planning systems. Furthermore, foundational models for service robots have been developed to enable more advanced autonomous movement. The ability of the service robots is tested in competitions such as RoboCup (in a home setting) and World Robot Summit (WRS) (in a convenience store setting).
As artificial intelligence continues to progress, so too does the performance of robotic vision systems and task planning. Computational methods tailored for service robots are vital, particularly since these units typically operate on battery power. There have been significant advancements in both large AI models suited for cloud computing and lightweight AI models designed for edge computing, thereby enhancing their functionality and efficiency.
Incorporating the latest technologies can enable the completion of tasks within a controlled environment. However, the development of robots for competitions and the social implementation of service robots presents challenges, especially when it comes to adapting to unfamiliar settings. In this organized session, we intend to discuss robot hardware, system integration, AI models, and its implimentation methods for service robots.
Topics
- Hardware design for service robots
- System integration for service robots
- Machine learning for service robots
- Generative AI for service robots
- Foundation model for service robots
- Embedded system for service robots
- Human-robot interaction
- Competition outcome from RoboCup and WRS
Important Dates
| Paper submission |
Paper Submission
Please submit your paper by following the paper submission procedure of ICIEV.
Accepted Papers
- Paper ID: 55, Title: Hand Gesture Recognition with Deep Learning-Free Feature Extraction
- Authors: Keita Okumura (Kyushu Institute of Tech.); Kosei Isomoto (Kyushu Institute of Tech.); Yuichiro Tanaka (Kyushu Institute of Tech.); Hakaru Tamukoh (Kyushu Institute of Tech.)
- Paper ID: 56, Title: Multi‑Speed Obstacle Detection Using Parallel SNN with Event‑Based Vision Sensors
- Authors: Yuta Ohno (Kyushu Institute of Tech.); Yuga Yano (Kyushu Institute of Tech.); Yuichiro Tanaka (Kyushu Institute of Tech.); Hakaru Tamkoh (Kyushu Institute of Tech.)
- Paper ID: 58, Title: Demonstration Data Quality Estimation with Modality Priority Adjustment Using Separate Encoders
- Authors: Hiromasa Yamaguchi (Kyushu Institute of Tech.); Yuga Yano (Kyushu Institute of Tech.); Hakaru Tamukoh (Kyushu Institute of Tech.)
- Paper ID: 59, Title: Number-Marked Prompting for Empty Space Estimation Considering Object Relationships in Unknown Environments
- Authors: Ryo Terashima (Kyushu Institute of Tech.); Yuga Yano (Kyushu Institute of Tech.); Koshun Arimura (Kyushu Institute of Tech.); Hakaru Tamkoh (Kyushu Institute of Tech.)
- Paper ID: 60, Title: Experimental Evaluation of Adaptive Dual-Mode Vision System for ROS2: Integrating YOLO-World and SAM2 for Robotic Manipulation
- Authors: GAI NAKATOGAWA (Tamagawa Univ.); SANSEI HORI (Tamagawa Univ.)
- Paper ID: 61, Title: Toolchain for Data Augmentation and Evaluation of Object Detection
- Authors: Ginga Kise (NIT, Kitakyushu); Yuma Yoshimoto (NIT, Kitakyushu)
- Paper ID: 62, Title: An FPGA Accelerated Architecture for Prolonging the Operating Time of Home Service Robots
- Authors: Haruki Miura (NIT, Kitakyushu); Rion Yofu (NIT, Kitakyushu); Yuma Yoshimoto (NIT, Kitakyushu)*
Organizers
Hakaru Tamukoh
Kyushu Institute of Technology
Yuichiro Tanaka
Kyushu Institute of Technology
Akinobu Mizutani
Kyushu Institute of Technology