The Robot Report Podcast · 20 Feb 2023 · 1h 23m
Mike Oitzman reports from the 2023 World Ag Expo, highlighting innovations from Agtonomy, Carbon Robotics, Farmwise, and Monarch Tractor. The episode details Agtonomy's 'Tele Farmer' remote-controlled tractor, Carbon Robotics' laser weeding implement, Farmwise's pivot to a smart implement, and Monarch's new Wingspan AI fleet management software. Hosts also discuss industry news including Zoox's autonomous vehicle testing and Tesla's FSD recall.
Where this conversation describes AI being used to build or deliver a product. Part of the AI in practice tracker.
Engineering & simulationAgtonomy
Agtonomy used an AI-based vision model called Trunk Vision to identify tree trunks and navigate safely around them without damage.
“they've got this, uh, feature they call trunk vision. Uh, and it's an AI-based vision model designed to identify trunks of trees and be able to navigate safely around them without damaging the tree trunks. So that's their their AI, uh, machine learning vision.”
Speaker 2 · Hear it at 29:32
Engineering & simulationFarmWisehigher frame per second capture
FarmWise upgraded its camera system to higher frame rate capture and developed custom actuators to improve the speed and precision of its machine learning-guided weed removal.
“we improve the cameras to the point where we have higher frame per second capture of images, which just increases the overall speed of all of the computations as well. But, and from there, we've developed our own actuator that's custom to this device, as well, that allows us to do a lot more precision actuation…”
Speaker 4 · Hear it at 45:31
Engineering & simulationCarbon Robotics
Carbon Robotics released a software upgrade enabling laser thinning, using vision models to identify crop spacing and kill extra plants while leaving beneficial ones.
“just announced, I think today, is that we now, on this module, just from a software upgrade, we now can do laser thinning. So along with your laser weeding, we can do laser thinnings. If you want to thin your crops out to stand, we can do that, all the way.”
Speaker 6 · Hear it at 1:04:20
Engineering & simulationMonarch Tractor (Wingspan AI)30 seconds before and after
Monarch Tractor's Wingspan AI platform uses computer vision to capture snapshots and video from multiple cameras, geotimestamping them to identify issues like crop disease or broken branches during autonomous operation.
“what we can actually do is take this little camera and create a snapshot. And what that snapshot does, is it captures, that's just a little nudification, it'll capture a picture from every camera on the tractor, and it'll geotimestamp those, so they know when and where that happened, and capture a video recording. And save a video recording from like 30 seconds before and after that snapshot.”
Speaker 5 · Hear it at 54:52
Engineering & simulationMonarch Tractor (Wingspan AI)
Monarch Tractor's autonomy system performs segmentation and classification on video data to support autonomous driving, generating insights when integrated with other farm data.
“it's not just that we're capturing the video data. It's that our autonomy system is going through, like, the very difficult challenge of delivering autonomous driving, and therefore, we are doing all types of segmentation and classification on the video that we see.”
Speaker 5 · Hear it at 59:40
early-stage companies and companies scaling up
Other shows on the same ground, closest first.
Jorge Heraud, VP of Automation and Autonomy at John Deere, explains how the company is leveraging IoT and computer vision to transform agriculture through products like See & Spray and fully autonomous tractors. He details the decision to refine existing herbicide application rather than pursuing unproven technologies like lasers, achieving a 3-5x reduction in chemical usage. Heraud also outlines John Deere's roadmap to a fully autonomous farm by 2030, highlighting the operational advantages of private land environments compared to public roads.
Daniel and Forbode discuss how robotic systems are revolutionizing agriculture by addressing inefficiencies in conventional farming. They analyze two key approaches: a configurable adaptive and scalable swarm system (CASS) from Texas A&M University, which uses collaborative drones and ground robots, and Pixel Farming Robotics, a Dutch company offering modular, lightweight autonomous robots. The hosts highlight specific technologies like wireless inductive charging from Wi-Fi On and Varda batteries, emphasizing how these innovations reduce soil compaction, chemical usage, and operational costs while increasing yield.
Gaurav Bansal, VP of Engineering at Blue River Technology (a subsidiary of John Deere), discusses the integration of IoT, AI, and computer vision in smart farming to address labor shortages and climate challenges. He details the John Deere autonomous tractor, which uses stereo cameras and NVIDIA Jetson GPUs for obstacle detection, and the See & Spray system, which uses real-time computer vision to reduce herbicide usage by up to two-thirds. Bansal also highlights the critical role of connectivity, noting a partnership with Starlink to solve rural coverage gaps, and outlines John Deere's goal of achieving fully autonomous production systems for corn and soybeans by 2030.
Hosts Matt Perrott (BuildPass) and Scott Wilcox (W Build) discuss the rapid evolution of AI agents, highlighting Google's new A2A protocol and OpenAI's acquisition talks with Codium. They analyze the competitive landscape between Tesla and Waymo in autonomous driving, the rise of drone delivery by Zipline, and the potential for humanoid robots to replace domestic labor while raising privacy concerns. The conversation also covers the decline of traditional consulting firms due to AI capabilities and the role of technology in rebuilding communities, such as post-disaster housing in LA.
Sophie Armanini of TU Munich and Imperial College London discusses bio-inspired aerial and aquatic robots, while Ben Scott-Robinson of Small Robot Company explains his team of autonomous farming robots (Tom, Dick, and Harry) designed for regenerative agriculture. Matthew Ryan Tucker from the University of Bristol details using a Boston Dynamics Spot robot for radiation mapping in the Chernobyl Exclusion Zone. The panel explores the challenges of deploying robots in wild, unpredictable environments, including weather, wildlife, and regulatory hurdles.
The summary, chapters and topics are generated from the episode’s audio by this directory, not written by The Robot Report Podcast. The episode itself belongs to its publisher.