Stanford Seminar - Where are the Field Robots?

December 6, 2024 Girish Chowdhary, UIUC Rapid advances in generative AI, reinforcement learning, and supervised learning have indeed transformed robotics across many industries. One domain that remains within reach, but still elusive, is when robots need to operate autonomously in harsh, dynamic, and unpredictable field environments over long durations. Cracking this domain is critical for solving some of the most pressing problems in sustainability, agriculture, and climate resilience. For instance, teams of autonomous robots hold the potential to address key challenges in modernizing agriculture. Yet, fully autonomous robots that operate without supervision for weeks, months, or for a whole growing season are not yet practical. In this talk I will outline a vision for the future of field robotics, highlighting key breakthroughs designed that advance robot AI without sacrificing reliability and practicality. I will explore advances in visual navigation, self-supervised learning, and robot onboard AI for increasing levels of autonomy. I will also discuss innovations in soft robotics for manipulation in cluttered environments. I will also share our progress in commercializing these technologies to benefit agriculture and solar energy industries through entrepreneurial activities. Including creating innovative products in high-throughput phenotyping, disease and pest monitoring, under-canopy cover crop planting for soil regeneration, automated spraying in orchard and tree crops for small and large farmers, and automating operations and maintenance in large solar farms. About the speaker: http://daslab.illinois.edu/ More about the course can be found here: https://stanfordasl.github.io/robotics_seminar/ View the entire AA289 Stanford Robotics and Autonomous Systems Seminar playlist: https://www.youtube.com/playlist?list=PLoROMvodv4rMeercb-kvGLUrOq4HR6BZD ► Check out the entire catalog of courses and programs available through Stanford Online: https://online.stan