IPAB Workshop - 05/12/2024

Title: Back to the Future of Integrated Robot Systems in the Era of Data-driven Models

 Abstract: Deep networks and "data-driven" models increasingly represent the state of the art for many problems in robotics. These methods and models are resource-hungry and opaque, and they are known to provide arbitrary decisions in previously unseen situations, whereas practical robotics applications  often require transparent, multi-step, multi-level decision-making and ad hoc collaboration under resource constraints and open world uncertainty. In this talk, I argue that for widespread use of robots, we need to revisit principles that can be traced back to the early pioneers of AI, and embed these principles in the architectures we develop for robots, with modern data-driven methods being just another tool in our toolbox. I will attempt to illustrate the potential benefits of this approach and outline a methodology for the design of such architectures.