By Chelsea Finn∗, Frederik Ebert∗, Sudeep Dasari, Annie Xie, Alex Lee, and Sergey Levine

With very little explicit supervision and feedback, humans are able to learn a wide range of motor skills by simply interacting with and observing the world through their senses. While there has been significant progress towards building machines that can learn complex skills and learn based on raw sensory information such as image pixels, acquiring large and diverse repertoires of general skills remains an open challenge. Our goal is to build a generalist: a robot that can perform many different tasks, like arranging objects, picking up toys, and folding towels, and can do so with many different objects in the real world without re-learning for each object or task.

By Esther Rolf∗, David Fridovich-Keil∗, and Max Simchowitz

In many tasks in machine learning, it is common to want to answer questions given fixed, pre-collected datasets. In some applications, however, we are not given data a priori; instead, we must collect the data we require to answer the questions of interest.

A crucial task for energy providers is the reliable and safe operation of their plants, especially when producing energy offshore. Autonomous mobile robots are able to offer comprehensive support through regular and automated inspection of machinery and infrastructure. In a world’s first pilot installation, transmission system operator TenneT tested the autonomous legged robot ANYmal on one of the world’s largest offshore converter platforms in the North Sea.

Researchers from EPFL and Stanford have developed small drones that can land and then move objects that are 40 times their weight, with the help of powerful winches, gecko adhesives and microspines.

By Daniel Seita, Jeff Mahler, Mike Danielczuk, Matthew Matl, and Ken Goldberg

This post explores two independent innovations and the potential for combining them in robotics. Two years before the AlexNet results on ImageNet were released in 2012, Microsoft rolled out the Kinect for the X-Box. This class of low-cost depth sensors emerged just as Deep Learning boosted Artificial Intelligence by accelerating performance of hyper-parametric function approximators leading to surprising advances in image classification, speech recognition, and language translation.

From about 245 to 66 million years ago, dinosaurs roamed the Earth. Although well-preserved skeletons give us a good idea of what they looked like, the way their limbs worked remains a bigger mystery. But computer simulations may soon provide a realistic glimpse into how some species moved and inform work in fields such as robotics, prosthetics and architecture.

By Xue Bin (Jason) Peng and Angjoo Kanazawa

Whether it’s everyday tasks like washing our hands or stunning feats of acrobatic prowess, humans are able to learn an incredible array of skills by watching other humans. With the proliferation of publicly available video data from sources like YouTube, it is now easier than ever to find video clips of whatever skills we are interested in.

The deployment of connected, automated, and autonomous vehicles presents us with transformational opportunities for road transport. These opportunities reach beyond single-vehicle automation: by enabling groups of vehicles to jointly agree on maneuvers and navigation strategies, real-time coordination promises to improve overall traffic throughput, road capacity, and passenger safety. However, coordinated driving for intelligent vehicles still remains a challenging research problem, and testing new approaches is cumbersome. Developing true-scale facilities for safe, controlled vehicle testbeds is massively expensive and requires a vast amount of space. One approach to facilitating experimental research and education is to build low-cost testbeds that incorporate fleets of down-sized, car-like mobile platforms.

In the future, smart textile-based soft robotic exosuits could be worn by soldiers, fire fighters and rescue workers to help them traverse difficult terrain and arrive fresh at their destinations so that they can perform their respective tasks more effectively. They could also become a powerful means to enhance mobility and quality of living for people suffering from neurodegenerative disorders and for the elderly.

Open-Source Software for robots is a de-facto standard in academia, and its advantages can benefit industrial applications as well. The worldwide ROS-Industrial initiative has been using ROS, the Robot Operating System, to this end.

Applications from 166 companies spread across 12 European countries and myriads of exiting robotics ideas was the beginning of the EU-funded initiative ROBOTT-NET in 2016.

In this post, we demonstrate how deep reinforcement learning (deep RL) can be used to learn how to control dexterous hands for a variety of manipulation tasks. We discuss how such methods can learn to make use of low-cost hardware, can be implemented efficiently, and how they can be complemented with techniques such as demonstrations and simulation to accelerate learning.

By John Miller

An earlier version of this post was published on Off the Convex Path. It is reposted here with the author’s permission.

In the last few years, deep learning practitioners have proposed a litany of different sequence models. Although recurrent neural networks were once the tool of choice, now models like the autoregressive Wavenet or the Transformer are replacing RNNs on a diverse set of tasks. In this post, we explore the trade-offs between recurrent and feed-forward models.

Three projects have made the final cut and received funding for a ROBOTT-NET pilot and even further development assistance.

Since programming is an extremely time-consuming business, small and medium-sized enterprises (SME) are often forced to manage without robots. Researchers from Fraunhofer IPA have therefore developed the software RobotKit specially for welding tasks. In an initial test scenario, the kit reduced programming times from 90 down to just 7 minutes.