Pepper as Crazy Eddie

When I was running RobotGalaxy, one of my investors commented that I needed a “barker” outside my stores to increase foot traffic.  At first I was confused, but then I quickly thought of the many barkers on 14th Street screaming at me to buy their insanely priced merchandise.  Yes, to many these seems like a good idea…

Earlier this week, SoftBank Group announced that its new Pepper robot will be available to businesses to fill the barker need, as a humanoid sales associate outside storefronts.  The company is developing a customer support package and a service that allows businesses to manage a fleet of Peppers, Fumihide Tomizawa, chief executive officer of SoftBank Robotics, said in an interview in Tokyo on Monday. The Japanese wireless carrier and Internet company investor will reveal more details of the service at the SoftBank World event July 30 and 31, he said. I guess this will be an early “Christmas in August” sale – “these prices are IN-SANE!”

When Pepper went on sale to Japanese consumers on June 20, all 1,000 units sold out within one minute. SoftBank is targeting businesses to increase the robot’s installed base which Tomizawa estimates would have to be in the hundreds of thousands to support a community of developers and an application store modeled on Apple Inc.’s.

“We see Pepper appearing on sales floors, behind reception desks, in educational and health care settings,” Tomizawa said. “Who buys what and under what circumstances — Pepper could analyze marketing data to modify his sales pitch.”


Pepper has already won over Mizuho Financial Group Inc., which said in March it will use the robot in some banks from this month and may roll it out to all branches across Japan. Nestle SA has plans to introduce Pepper to 1,000 stores in the country by the end of the year to help out with sales.

The robot costs 198,000 yen ($1,600) and comes with an optional 14,800 yen monthly service plan giving users access to cloud-based voice-recognition and an app store. SoftBank is readying a separate plan for business users, Tomizawa said.

Featuring more than 20 motors and highly articulated arms, Pepper is capable of human-like body language. Its shoulders heave when in standby mode, imitating sleep. But it’s not designed for menial tasks. Instead, SoftBank founder Masayoshi Son is betting the robot’s friendly physical appearance will spur adoption of cloud services and attract app developers.

Son is directly in charge of the robot business and has said it is central to his 30-year vision for Japan’s third-largest wireless carrier. Son has taken an active role in creating Pepper, weighing in on decisions ranging from the use of wheels instead of legs to the minute details of its body design, according to Tomizawa.

The attention has Tomizawa fielding inquiries from Son, Japan’s second-richest man, including a 4 a.m. phone call asking him to come up with a list of possible patents by noon. Son has said he personally has applied for several patents related to Pepper.

“Son always brings full attention to every business he’s involved in, but it’s even more so with robotics,” said Tomizawa, who’s been with SoftBank for 15 years. “He’s really serious about this.”

While I have no doubt that Pepper will be successful in raising the curiosity of shoppers, I just think it is an incredible waste of resources that could be better applied to helping our most vulnerable citizens cope with terminal illness or aging.

Please note that over the next three weeks RobotRabbi will be taking a summer vacation…

Mind Controlled Robots

The first episode of Star Trek, The Menagerie, Captain Christopher Pike (not Kirk) lands on a distant planet and falls in love with a girl (Vina) that is captured by aliens that communicate telepathically.  Pike attempts to rescue her, but quickly learns that her beauty is really only an illusion put on by the aliens.  Later on in the episode, Pike is injured and becomes as deformed as Vina. Now, imprisoned in a wheelchair, the Captain returns to the planet to reunite with his true love under the guise of the aliens’ ruse. Trek-Menagerie29 A team of researchers at the Swiss Federal Institute of Technology’s Defitech Foundation Chair in Brain-Machine Interface in Lausanne, Switzerland, are working on creating the ultimate joystick –  a brain-to-computer interface that could enable disabled people to control robots with their minds.  According to the project lead,Professor José del R. Millán, “we have been developing brain-computer interfaces for people who suffer different kinds of motor disabilities so that they can translate their mental intentions into commands for the robots.”

  The project, which has been underway for a year, has tested the interface with nine people who are disabled and 10 healthy people across Italy, Germany and Switzerland. First, the user has to train to communicate with the robot. A certain thought will light up an area of the brain. To turn the robot left and right, the user will have to think in a specific way. These electrical brain signals are picked up by a non-invasive cap fitted with electrodes. After training how to use the cap, the users then controlled a telepresence robot on location in a laboratory in Switzerland while still in their homes, sometimes in different countries, in real-time. The robot, still in its early stages, consists of a laptop on a wheeled frame. This has a camera that allows the user to see the environment around the robot, as well as a display that shows the user’s face via Skype, letting the user have conversations with people at the robot’s location (imagine if this technology could work with Vgo or iRobot’s Cisco telepresence device). Additionally, the robot is fitted with sensors that detect the proximity of objects in the room — allowing it to avoid collisions on its own, without being micromanaged by the user. So far, the laboratory is reporting a 100 percent success rate. Users were capable of directing the robot from room to room since the first trial, said Professor Millán. “But then we went over to compare their performance against 10 people without any kind of motor disabilities, and we saw that their performance was essentially the same.” Each of the users who are disabled were able to easily control the telepresence robot with less than 10 days of training. human_computer5 The project, part of the European Commission-funded Tools for Brain-Computer Interface project, is still in the testing phase, and it’s yet to be made available to users. The team hopes that the technology will be made available to the public, but there are some issues that need to be overcome. “We would like to see this technology at the user’s site, not confined to the laboratory,” Professor Millán said. “For this to happen, insurance companies will have to help finance these technologies.”

Unlike Star Trek, this telepathic control is not an illusion but a real quality of life improvement that arguments ones disabilities with robots.

Robotic Bridge Builders

Amsterdam is a city of bridges, 1,500 in fact, and ripe for disruption. As a follow up to my story last May about robots in architecture, I am pleased to introduce the new bridge building machine, the MX3D.


Dutch designer Joris Laarman has designed a pedestrian bridge for Amsterdam that will be 3D printed by robots. The ornate metal structure, which will span a canal in the Dutch city, will be printed in-situ by robotic arms. The location of the bridge will be announced soon and completion is set for 2017.

3D printed bridge by Joris Laarman and MX3D

The versatile six-axis robots – which are able to rotate their arms along six different planes of movement – will print a load-bearing structure that will support their own weight as they work. This will allow them to start on one bank of the canal and work their way across to the other side, printing steel as they go.


“This bridge will show how 3D printing finally enters the world of large-scale, functional objects and sustainable materials while allowing unprecedented freedom of form,” said Laarman. “The symbolism of the bridge is a beautiful metaphor to connect the technology of the future with the old city, in a way that brings out the best of both worlds.”

Joris Laarman 3D printed bridge

The project has been developed by MX3D, a technology startup launched by Joris Laarman Lab to investigate ways of printing large, sophisticated structures.

It builds on technology developed by Laarman that allows industrial robots to “draw” metal structures in the air. This potentially allows far larger structures to be printed than are currently possible, and means the technology can start to move out of the factory and onto the construction site.

“What distinguishes our technology from traditional 3D printing methods is that we work according to the ‘printing outside the box’ principle,” said Tim Geurtjens, chief technology officer at MX3D.

Joris Laarman 3D printed bridge

“By printing with six-axis industrial robots, we are no longer limited to a square box in which everything happens,” Geurtjens added. “Printing a functional, life-size bridge is of course the ideal way to showcase the endless possibilities of this technique.”

The project, which is supported by design software company Autodesk and construction company Heijmans, could eventually allow entire buildings to be printed.

“The MX3D platform is a potential game changer,” said Maurice Conti, director of strategic innovation at Autodesk. “Breaking free of the traditional limitations of additive manufacturing – small-size prints and poor material performance – this technology opens up possibilities for architectural-scale, relatively low-cost, metal structures that are as complex as the designer’s imagination.”

Laarman is a pioneering designer who works at the interface between technology and traditional object-making. Projects by the 35-year-0ld include 3D-printed metal chairs and explorations in making furniture bades on the way bones grow.

The MX3D is part of a new wave of robotic assisted building machines, however if Laarman accomplishes his machine it will be the first invention in the construction industry to run on autopilot.

And the winner is…

Last Saturday, everyone waited with bated breadth to see if American Pharoah would win the Triple Crown.  At the same time, under the California sun, 23 teams competed in the ultimate robotic contest – the DARPA’s Robotic Challenge (DRC).


It should come as no surprise that a South Korean team won the $2 million grand prize, as South Korea is the number one country for robotic deployments. Team KAIST’s bipedal bot accepted the prize while scooting around the venue on its wheel knees. The winning design managed to navigate DARPA’s obstacle course in under 45 minutes, successfully completing eight natural disaster-related tasks including walking over rubble, driving a car, tripping circuit breakers, and turning valves.

the winner

The DRC was set up after the 2011 Fukushima nuclear disaster in Japan, with the aim of accelerating the development of robots that can respond to man-made or natural disasters. Twenty-three teams competed in the finals, with a dozen entering from the US, and the rest traveling from countries including Germany, Italy, Japan, and Hong Kong.

The teams had been developing their robots for more than two years and have tried the challenges before. However, while previous trials gave the robots 30 minutes to complete each task, the two-day finals — staged in front of thousands of spectators in California — pushed the teams to complete all eight in less than an hour. Only three robots managed to successfully tackle them all, and the rest, well, they fell down a lot (see blooper reel).


“These robots are big and made of lots of metal, and you might assume people seeing them would be filled with fear and anxiety,” said the event’s organizer Gill Pratt in a press statement. “But we heard groans of sympathy when those robots fell. And what did people do every time a robot scored a point? They cheered! It’s an extraordinary thing, and I think this is one of the biggest lessons from DRC — the potential for robots not only to perform technical tasks for us, but to help connect people to one another.”

Although Team KAIST took home the top prize, second place (and $1 million) went to Team IHMC Robotics, who used the Atlas robot built by Google-owned Boston Dynamics. Third place and $500,000 went to Team Tartan Rescue from Carnegie Mellon University, who finished with the top score at the end of the first day but were eventually overtaken. American wins two!

team kaist

“This is the end of the DARPA Robotics Challenge but only the beginning of a future in which robots can work alongside people to reduce the toll of disasters,” said DARPA director Arati Prabhakar in a press statement. “I am so proud of all the teams that participated and know that the community that the DRC has helped to catalyze will do great things in the years ahead.”


Maybe the next Triple Crown winner will be American Bot….

Origami Robots Begin Their Fantastic Voyage

While speaking with an innovative robotic company this week, I concluded it takes a lot of humans to sell just one robot.  The employment around robotics was further validated by Sunday’s WSJ article about Uber poaching scientists from Carnegie Mellon University.  While there will be job fallout, robots in the long run will open up the field to new unforeseen levels, even microscopic…

Researchers from the Massachusetts Institute of Technology (MIT) and the Technische Universität München (TU Munich) in Germany have unveiled a small origami robot that can fold itself, walk around, and even self-destruct. Experts believe this new technology can be used to help doctors perform key medical operations from within the patient’s body (think the Fantastic Voyage)

The presentation was made earlier this week at the International Conference on Robotics and Automation (ICRA) in Seattle, Washington hosted by Institute of Electrical and Electronics Engineers.

Designed and built by scientists from MIT’s Computer Science and Artificial Intelligence Laboratory, the tiny robot measures at around 1.7 centimeters long and weighs about one third of a gram.

The robot’s primary feature is a neodymium magnet layered with polystyrene and PVC. When heated, the material allows it to take on another form similar to an origami.

A multi-stage folding process can also be achieved by heating the robot at varying degrees. Low heat can give the shape-shifting robot one form, but placing it back on the heating pad and raising the temperature can result in a secondary design.

Instead using a set of wires to make the robot move, it uses an external magnetic field made up of four coils to guide it along its path. These magnets power its movement by cycling on and off at 15 hertz.


The MIT robot can be made to walk or roll depending on its task, and it can also be disintegrated by exposing it to acetone. Its designers are also experimenting on the robot’s structural material to allow it to be dissolved in water.

The researchers now plan to create another version of the shape-shifting robot with onboard sensors to allow it to operate on its own. The hope is that the robot could one day be used for medical purposes.

A smaller robot could be inserted into an individual’s body, navigate through the bloodstream, perform medical tasks and then eventually dissolve in the stomach.

The technology could also be used for planetary explorations. The robot could be programmed to collect various soil samples or even journey through small cracks in the ground.

Origami has been around for close to 2,000 years, now as we enter the new age of connected devices, we see innovations impact us in novel and dramatic ways.