Research Project
Collaborative Robot Field Management and Standardized Verification Platform
This project aims to establish internationally standardized testing environments, ensuring safety and efficiency in human-robot collaboration.
Builds a collaborative robot field management and verification platform integrating 5G/6G, AI-RAN, and LLM with ISO-certified safety standards.
Research Project
This project aims to establish internationally standardized testing environments, ensuring safety and efficiency in human-robot collaboration.
Principal Investigator
En-Cheng Liou Assistant Professor
Research Field: Integrated Sensing and Communication (ISAC) for 6G, Open Radio Access Network (O-RAN), Information and Communication Security, Artificial Intelligence of Things (AIoT), Indoor Localization, Wireless Network Softwarization and Virtualization
PI DetailsResearch Project
This sub-project is responsible for building the "Collaborative Robot Field Management and Standardized Verification Platform," which brings 5G/6G communications, AI-RAN, and Large Language Models (LLM) into the collaboration verification flow of humanoid robots, providing shared testing and deployment infrastructure for other sub-projects working on healthcare or smart-manufacturing applications. The technical work proceeds along two main tracks. The first track is AI-adaptive wireless localization. Robots first build the field fingerprint quickly, then keep collecting WiFi, 5G, BLE, and magnetic-field signals during routine tasks and feed them back into the localization model in a continuously optimizing closed loop, so that accuracy improves with usage and achieves the "improving-with-use" effect. The second track is the international standard certification system. The four standards, ISO 42001 (AI management), ISO 27001 (information security), ISO 27701 (privacy) and ISO 17025 (laboratory measurement), are introduced into the development and operations workflow of the robot system, so that the system has a compliant and traceable basis when deployed in clinical environments. On top of these, this sub-project also builds the Agent Operation unified service platform, which provides code version control, Prompt Registry and Model Registry management, and automated CI/CD deployment to Kubernetes clusters for the AI Agents of every sub-project, so that each sub-project does not have to build its own operations infrastructure.
Principal Investigator
Ta-Sung Lee Distinguished Professor
Research Field: Wireless Communication Technology, Intelligent Signal Processing, Wireless Sensing Technology, Smart Networks, Radio Resource Management
PI DetailsResearch Project
This work proposes a cloud-native, multi-agent–based platform for managing and orchestrating heterogeneous robots. Centered on a Coordinator Agent, the system integrates real-time robot perception and remote-control capabilities and follows the Agent-to-Agent and Model Context Protocol standards for task parsing, assignment, and monitoring. The Coordinator Agent is responsible for task planning and coordination, dispatching Robot Agents to execute tasks. Robot Agents invoke tools to control robots for task execution and status reporting, while the Coordinator Agent can further coordinate multiple robots based on feedback to accomplish complex tasks. In addition, the platform incorporates AgentOps, MLOps, and automated deployment with Kubernetes, supporting version management, automated testing, continuous integration, and continuous deployment for Agents, Prompts, Models, and Tools, as well as model training and deployment. Overall, the system enhances the efficiency, flexibility, and scalability of robot collaboration, and its feasibility is validated through practical implementation.
Principal Investigator
Chien-Chao Tseng Distinguished Professor
Research Field: Software Defined Networking、Network Function Virtualization, NFV-MANO and Acceleration Techniquefor B5G/6G Networks, CICD, DevOps, and Cloud-Native Technologies, Agentis AI and AgentOps、Satellite Network Management
PI Details