Video Friday: Lifelike Robot Heads, and More

Your weekly selection of awesome robot videos

5 min read

Erico Guizzo is IEEE Spectrum's Digital Innovation Director.

Mesmer robot heads
Image: Engineered Arts via YouTube

Video Friday is your weekly selection of awesome robotics videos, collected by your Automaton bloggers. We’ll also be posting a weekly calendar of upcoming robotics events for the next few months; here’s what we have so far (send us your events!):

ELROB 2018 – September 24-28, 2018 – Mons, Belgium
ARSO 2018 – September 27-29, 2018 – Genoa, Italy
ROSCon 2018 – September 29-30, 2018 – Madrid, Spain
IROS 2018 – October 1-5, 2018 – Madrid, Spain
Japan Robot Week – October 17-19, 2018 – Tokyo, Japan
ICSR 2018 – November 28-30, 2018 – Qingdao, China

Let us know if you have suggestions for next week, and enjoy today’s videos.

Nothing creepy about this, nothing at all.

Built by Engineered Arts, 2 Mesmer Heads perform a synchronised sequence. One is complete with lifelike skin and hair, the other is showing it's mechanical workings.

Engineered Arts ]

As a joint effort between the CARIS lab and LCI at UBC, we have programmed our PR2 to navigate between the two laboratories located in different floors by taking the elevator. The ultimate goal is to eventually get PR2 to deliver mail and packages between the two locations.

Elevator riding with a robot as big as the PR2 gets complicated when other humans are involved, but the robot is occasionally polite enough to let people go first. Occasionally.

[ CARIS Lab ]

These shoes use pneumatic McKibben muscles to snug themselves down instead of laces:

Here’s what I want: Shoes just like these, except with a pump in the heel actuated by your weight. So, slip the shoes on, and they tighten as you walk, with an adjustable valve to let you set the maximum pressure. Also whatever that robotic gauntlet thing is, I want one of those, too.

[ Suzumori Endo Lab ]

Lockheed Martin and Drone Racing League announced an innovation competition, challenging teams to develop artificial intelligence (AI) technology that will enable an autonomous drone to race a pilot-operated drone – and win. Participating teams will compete in a series of challenges for their share of over $2 million in prizes.

The AlphaPilot challenge aims to accelerate the development and testing of fully autonomous drone technologies. AlphaPilot participants will design an artificial intelligence/machine learning framework, powered by the NVIDIA Jetson platform for AI at the edge, capable of flying a drone – without any pre-programming or human intervention – through challenging multi-dimensional race courses in DRL’s new Artificial Intelligence Robotic Racing Circuit.

[ AlphaPilot ]

Want an Aibo but can’t afford an Aibo? Takara Tomy has the next best thing, and it’s only US $135:

Uh, isn’t avoiding the time-to-go-pee behavior the whole reason to get a robotic dog in the first place?

[ Takara Tomy ]

This is a pretty good (albeit short) look at Boeing’s forthcoming MQ-25 unmanned carrier-based tanker drone.

[ Boeing ]

This Cozmo commercial is even better when you have no idea what they’re saying.

[ Cozmo ]

This is a demonstration of a robot’s abrupt stopping motions (typical of industrial collision avoidance mechanisms) and a novel human-like hesitation motions designed at the CARIS lab. The experimenter is intentionally triggering the two types of collision avoidance. The human-like designed gestures are perceived communicative of hesitation to human observers.

[ CARIS Lab ]

The video introduces the adult-sized open-source 3D printed humanoid robot NimbRo-OP2X, which was developed by the Autonomous Intelligent Systems group of University of Bonn, Germany.

[ NimbRo ]

The Ishikawa Watanabe Laboratory is all about doing interesting stuff by combining very high speed robots with even higher speed cameras. Here’s a sample of their past work, including at least one clip I haven’t seen before.

[ University of Tokyo ]

Today at the D60 symposium, researchers supported by DARPA’s Fast Lightweight Autonomy (FLA) program demonstrated autonomous flight by a small unmanned aerial vehicle equipped with cameras and sensors and guided by advanced autonomy algorithms. That software runs a small quadcopter (~5 lbs.) with limited battery power and onboard computer processing. As shown in this video, the FLA system can make its way through a new environment with no prior knowledge of the space, navigating to and from waypoints set by its operator and independently avoiding obstacles in its path.

FLA’s algorithms have been demonstrated so far on air vehicles only, but they could be used on small, lightweight ground vehicles as well.

[ DARPA ]

ROBOTT-NET: The people behind the person behind the robot.

But who are the people behind the people behind the person behind the robot behind the person, if the ROBOTT-NET folks aren’t sure how to teach a robot to sneak up on someone?

[ ROBOTT-NET ]

Thanks Thilo!

Good news for DJI M200 series drone owners: your drone may survive getting caught in a propane explosion.

[ DJI ]

As someone who used to work in a factory doing jobs very similar to this, it’s always nice to see that the robots (like Sawyer below) are taking over.

[ Rethink ]

eBee X is the fixed-wing drone for all your mapping needs. Designed to boost the quality, efficiency and safety of your data collection, it has a camera to suit every job, the accuracy and coverage to meet every project’s requirements, and can work virtually every type of site.

[ SenseFly ]

In which Roboy does my job for me:

In this video we discuss the paper: "Joint Optimization of Robot Design and Motion Parameters using the Implicit Function Theorem" by Sehoon Ha, Stelian Coros, Alexander Alspach, Joohyung Kim and Katsu Yamane - all belong to Disney Research, USA. What is an optimal way to design a robots structure? Can an algorithm help us in designing and optimizing robot morphology? We discuss all this here. Note: We are reviewing this paper, and we were not part of the research or the team which authored this paper.

[ Disney Research ] via [ Roboy ]

As NASA celebrates its 60th anniversary, this panel discussion will look back over the decades at how far our robotic exploration has come, and consider where we might be headed.

Much has changed about the way we explore space in the 60 years since NASA began operations on Oct. 1, 1958. Today’s robotic spacecraft are beginning to experiment with laser communications, artificial intelligence and 3-D printed parts. But some of the first spacecraft the U.S. sent to the Moon included parts made of wood; some used to record data on motorized magnetic tape recorders. Despite all the advances, one thing hasn’t changed: we still rely on robotic spacecraft to extend our senses above and beyond Earth and to blaze a trail as precursors for human explorers.

[ Caltech ]

The Conversation (0)