Cloud and Fog Robotics in the Age of Deep Robot Learning

Organizers: Sandeep P. Chinchali, Ajay Tanwani, Marco Pavone


Cloud and fog robotics enables resource-constrained robots to utilize both on-robot and cloud resources for compute and storage. Recent progress in deep learning, cloud computing, and the advent of cloud robotics platforms from the likes of Google and Amazon make today an exciting time to consider cloud robotics.

However, cloud robotics comes with an often understated cost: offloading images, videos or high data-rate sensor measurements such as LIDAR can severely congest wireless networks, add to latency, and place a large burden on cloud compute resources. In this workshop, we aim to bridge advances in the computer systems and robotics communities to address how to best distribute networking, storage, and communication resources between robots and the cloud.

Accordingly, the objectives of this workshop are:

  1. To convene together researchers and industry experts from computer science, systems, robotics and deep learning to jointly discuss the challenges and define a roadmap for technology development.
  2. To identify promising applications of cloud robotics, such as offloading object detection, grasp planning, motion planning, mapping and localization for robots to enhance dexterous manipulation.
  3. To inform roboticists about algorithms for network-enabled services and introduce new cloud robotics platforms from Amazon, Google, and other university partners.