AI-Driven Task & Motion Planning for Service Robots with Oracle Cloud Infrastructure
Published in Oracle AI & Data Science Blog, 2025
This article details a joint pilot project between SocRob@Home (ISR-IST) and Oracle, initiated following the 2024 edition of JEEC (Electrical and Computer Engineering Week) at Instituto Superior Técnico.
Building General Purpose Service Robots (GPSR) capable of processing natural language instructions requires significant computational power for fine-tuning Large Language Models (LLMs) and orchestrating complex decision-making pipelines. In this work, we evaluated Oracle Cloud Infrastructure (OCI) to power our LLM-driven task and motion planning framework for domestic robots.
By leveraging OCI’s GPU-powered instances and Generative AI tooling, we demonstrated how high-performance cloud infrastructure can streamline LLM fine-tuning, action tool execution, and context-aware task generation. This collaboration highlights how cloud-accelerated AI pipelines can scale real-world service robotics, making autonomous domestic systems more interactive, precise, and adaptable.
Read the full publication on the Oracle AI & Data Science Blog.
Recommended citation: P. Bordalo-Machado, A. Certo, R. Serra, R. Coimbra, H. Jain, and D. Soni. "AI-Driven Task & Motion Planning for Service Robots with Oracle Cloud Infrastructure," Oracle AI & Data Science Blog. (Dec. 2025), url: https://blogs.oracle.com/ai-and-datascience/ai-driven-task-motion-planningservice-robots-with-oci.
Download Paper