In August, over 350 new datasets were published on Kaggle, in part sparked by our $10,000 Datasets Publishing Award. This interview delves into the stories and background of August’s three winners–Ugo Cupcic, Sudalai Rajkumar, and Colin Morris. They answer questions about what stirred them to create their winning datasets and kernel ideas they’d love to see other Kagglers explore.
Our team selected this dataset because it combines two exciting fields of research: robotics and deep learning. To learn more about the dataset, we loved Ugo’s excellent blog post “How I taught my robot to realize how bad it was at holding things” complementing the usage of Kaggle to make such a unique research-based dataset open and accessible to all.
As the Chief Technical Architect of the Shadow Robot Company, I spend a lot of time thinking about grasping things with our robots. This story is a quick delve into the world of grasp robustness prediction using machine learning.
Getting started with robotics is probably a lot easier than you think. Here’s a simulation sandbox that’s cross-platform and provides a simple high-level API. It should help you get started experimenting with robot grasping tasks.
As the Chief Technical Architect at the Shadow Robot Company, I spend a lot of time playing with different algorithms to see how they’d fit our robots. Controlling a complex robot to make it behave the way you’d want in a complex environment is… complex!
A robot Hand without a robot Arm is most of the time useless. At Shadow we have a long history of interfacing different robot arms with our software and hardware. In the different projects we’ve run over the years, we’ve written software for arms from Universal Robot, Denso, Kuka, Staubli… We’ve also developed a few intriguing arms internally, from an arm actuated by air muscles to a lightweight arm that picks-up strawberries.
On this journey, we’ve learned a few things. Let me share a few tips on what it takes to write a good interface for a robot arm quickly.
At Shadow, we’re focusing on making complex robots intuitive to use. For that, we need very good path planning. There are plenty of amazing solutions out there, but we were recently faced with a project where those state of the art solutions just weren’t good enough for us. We needed a super fast planner that generated trajectories that “looked good”.
An important part of my job is to stay on top of what’s currently happening in robotics. Given the fast pace in robotics, it sometimes feels more like trying not to drown! A great way to see all the latest trends is to attend a few key conferences. IROS and ICRA are the two biggest robotics conferences in the world and I often go to those. I attend a few other smaller more focused conferences to study precise subjects – such as control theory at the great DLMC conference in Zurich.
An important part of our roadmap is focusing on making grasping trivial for the end user. We want to be able to point our robot at an object with the instruction of grasp it. Although from a human point of view it sounds trivial, this is actually complicated for a robot. A crucial step in that direction is to be able to quantify how well the robot is grasping the object; without that measurement, the robot will never be able to improve. In this post, we’ll focus on different methods used to assess grasp quality.