Tag Archives: making

#436190 What Is the Uncanny Valley?

Have you ever encountered a lifelike humanoid robot or a realistic computer-generated face that seem a bit off or unsettling, though you can’t quite explain why?

Take for instance AVA, one of the “digital humans” created by New Zealand tech startup Soul Machines as an on-screen avatar for Autodesk. Watching a lifelike digital being such as AVA can be both fascinating and disconcerting. AVA expresses empathy through her demeanor and movements: slightly raised brows, a tilt of the head, a nod.

By meticulously rendering every lash and line in its avatars, Soul Machines aimed to create a digital human that is virtually undistinguishable from a real one. But to many, rather than looking natural, AVA actually looks creepy. There’s something about it being almost human but not quite that can make people uneasy.

Like AVA, many other ultra-realistic avatars, androids, and animated characters appear stuck in a disturbing in-between world: They are so lifelike and yet they are not “right.” This void of strangeness is known as the uncanny valley.

Uncanny Valley: Definition and History
The uncanny valley is a concept first introduced in the 1970s by Masahiro Mori, then a professor at the Tokyo Institute of Technology. The term describes Mori’s observation that as robots appear more humanlike, they become more appealing—but only up to a certain point. Upon reaching the uncanny valley, our affinity descends into a feeling of strangeness, a sense of unease, and a tendency to be scared or freaked out.

Image: Masahiro Mori

The uncanny valley as depicted in Masahiro Mori’s original graph: As a robot’s human likeness [horizontal axis] increases, our affinity towards the robot [vertical axis] increases too, but only up to a certain point. For some lifelike robots, our response to them plunges, and they appear repulsive or creepy. That’s the uncanny valley.

In his seminal essay for Japanese journal Energy, Mori wrote:

I have noticed that, in climbing toward the goal of making robots appear human, our affinity for them increases until we come to a valley, which I call the uncanny valley.

Later in the essay, Mori describes the uncanny valley by using an example—the first prosthetic hands:

One might say that the prosthetic hand has achieved a degree of resemblance to the human form, perhaps on a par with false teeth. However, when we realize the hand, which at first site looked real, is in fact artificial, we experience an eerie sensation. For example, we could be startled during a handshake by its limp boneless grip together with its texture and coldness. When this happens, we lose our sense of affinity, and the hand becomes uncanny.

In an interview with IEEE Spectrum, Mori explained how he came up with the idea for the uncanny valley:

“Since I was a child, I have never liked looking at wax figures. They looked somewhat creepy to me. At that time, electronic prosthetic hands were being developed, and they triggered in me the same kind of sensation. These experiences had made me start thinking about robots in general, which led me to write that essay. The uncanny valley was my intuition. It was one of my ideas.”

Uncanny Valley Examples
To better illustrate how the uncanny valley works, here are some examples of the phenomenon. Prepare to be freaked out.

1. Telenoid

Photo: Hiroshi Ishiguro/Osaka University/ATR

Taking the top spot in the “creepiest” rankings of IEEE Spectrum’s Robots Guide, Telenoid is a robotic communication device designed by Japanese roboticist Hiroshi Ishiguro. Its bald head, lifeless face, and lack of limbs make it seem more alien than human.

2. Diego-san

Photo: Andrew Oh/Javier Movellan/Calit2

Engineers and roboticists at the University of California San Diego’s Machine Perception Lab developed this robot baby to help parents better communicate with their infants. At 1.2 meters (4 feet) tall and weighing 30 kilograms (66 pounds), Diego-san is a big baby—bigger than an average 1-year-old child.

“Even though the facial expression is sophisticated and intuitive in this infant robot, I still perceive a false smile when I’m expecting the baby to appear happy,” says Angela Tinwell, a senior lecturer at the University of Bolton in the U.K. and author of The Uncanny Valley in Games and Animation. “This, along with a lack of detail in the eyes and forehead, can make the baby appear vacant and creepy, so I would want to avoid those ‘dead eyes’ rather than interacting with Diego-san.”

​3. Geminoid HI

Photo: Osaka University/ATR/Kokoro

Another one of Ishiguro’s creations, Geminoid HI is his android replica. He even took hair from his own scalp to put onto his robot twin. Ishiguro says he created Geminoid HI to better understand what it means to be human.

4. Sophia

Photo: Mikhail Tereshchenko/TASS/Getty Images

Designed by David Hanson of Hanson Robotics, Sophia is one of the most famous humanoid robots. Like Soul Machines’ AVA, Sophia displays a range of emotional expressions and is equipped with natural language processing capabilities.

5. Anthropomorphized felines

The uncanny valley doesn’t only happen with robots that adopt a human form. The 2019 live-action versions of the animated film The Lion King and the musical Cats brought the uncanny valley to the forefront of pop culture. To some fans, the photorealistic computer animations of talking lions and singing cats that mimic human movements were just creepy.

Are you feeling that eerie sensation yet?

Uncanny Valley: Science or Pseudoscience?
Despite our continued fascination with the uncanny valley, its validity as a scientific concept is highly debated. The uncanny valley wasn’t actually proposed as a scientific concept, yet has often been criticized in that light.

Mori himself said in his IEEE Spectrum interview that he didn’t explore the concept from a rigorous scientific perspective but as more of a guideline for robot designers:

Pointing out the existence of the uncanny valley was more of a piece of advice from me to people who design robots rather than a scientific statement.

Karl MacDorman, an associate professor of human-computer interaction at Indiana University who has long studied the uncanny valley, interprets the classic graph not as expressing Mori’s theory but as a heuristic for learning the concept and organizing observations.

“I believe his theory is instead expressed by his examples, which show that a mismatch in the human likeness of appearance and touch or appearance and motion can elicit a feeling of eeriness,” MacDorman says. “In my own experiments, I have consistently reproduced this effect within and across sense modalities. For example, a mismatch in the human realism of the features of a face heightens eeriness; a robot with a human voice or a human with a robotic voice is eerie.”

How to Avoid the Uncanny Valley
Unless you intend to create creepy characters or evoke a feeling of unease, you can follow certain design principles to avoid the uncanny valley. “The effect can be reduced by not creating robots or computer-animated characters that combine features on different sides of a boundary—for example, human and nonhuman, living and nonliving, or real and artificial,” MacDorman says.

To make a robot or avatar more realistic and move it beyond the valley, Tinwell says to ensure that a character’s facial expressions match its emotive tones of speech, and that its body movements are responsive and reflect its hypothetical emotional state. Special attention must also be paid to facial elements such as the forehead, eyes, and mouth, which depict the complexities of emotion and thought. “The mouth must be modeled and animated correctly so the character doesn’t appear aggressive or portray a ‘false smile’ when they should be genuinely happy,” she says.

For Christoph Bartneck, an associate professor at the University of Canterbury in New Zealand, the goal is not to avoid the uncanny valley, but to avoid bad character animations or behaviors, stressing the importance of matching the appearance of a robot with its ability. “We’re trained to spot even the slightest divergence from ‘normal’ human movements or behavior,” he says. “Hence, we often fail in creating highly realistic, humanlike characters.”

But he warns that the uncanny valley appears to be more of an uncanny cliff. “We find the likability to increase and then crash once robots become humanlike,” he says. “But we have never observed them ever coming out of the valley. You fall off and that’s it.” Continue reading

Posted in Human Robots

#436188 The Blogger Behind “AI ...

Sure, artificial intelligence is transforming the world’s societies and economies—but can an AI come up with plausible ideas for a Halloween costume?

Janelle Shane has been asking such probing questions since she started her AI Weirdness blog in 2016. She specializes in training neural networks (which underpin most of today’s machine learning techniques) on quirky data sets such as compilations of knitting instructions, ice cream flavors, and names of paint colors. Then she asks the neural net to generate its own contributions to these categories—and hilarity ensues. AI is not likely to disrupt the paint industry with names like “Ronching Blue,” “Dorkwood,” and “Turdly.”

Shane’s antics have a serious purpose. She aims to illustrate the serious limitations of today’s AI, and to counteract the prevailing narrative that describes AI as well on its way to superintelligence and complete human domination. “The danger of AI is not that it’s too smart,” Shane writes in her new book, “but that it’s not smart enough.”

The book, which came out on Tuesday, is called You Look Like a Thing and I Love You. It takes its odd title from a list of AI-generated pick-up lines, all of which would at least get a person’s attention if shouted, preferably by a robot, in a crowded bar. Shane’s book is shot through with her trademark absurdist humor, but it also contains real explanations of machine learning concepts and techniques. It’s a painless way to take AI 101.

She spoke with IEEE Spectrum about the perils of placing too much trust in AI systems, the strange AI phenomenon of “giraffing,” and her next potential Halloween costume.

Janelle Shane on . . .

The un-delicious origin of her blog
“The narrower the problem, the smarter the AI will seem”
Why overestimating AI is dangerous
Giraffing!
Machine and human creativity

The un-delicious origin of her blog IEEE Spectrum: You studied electrical engineering as an undergrad, then got a master’s degree in physics. How did that lead to you becoming the comedian of AI?
Janelle Shane: I’ve been interested in machine learning since freshman year of college. During orientation at Michigan State, a professor who worked on evolutionary algorithms gave a talk about his work. It was full of the most interesting anecdotes–some of which I’ve used in my book. He told an anecdote about people setting up a machine learning algorithm to do lens design, and the algorithm did end up designing an optical system that works… except one of the lenses was 50 feet thick, because they didn’t specify that it couldn’t do that.
I started working in his lab on optics, doing ultra-short laser pulse work. I ended up doing a lot more optics than machine learning, but I always found it interesting. One day I came across a list of recipes that someone had generated using a neural net, and I thought it was hilarious and remembered why I thought machine learning was so cool. That was in 2016, ages ago in machine learning land.
Spectrum: So you decided to “establish weirdness as your goal” for your blog. What was the first weird experiment that you blogged about?
Shane: It was generating cookbook recipes. The neural net came up with ingredients like: “Take ¼ pounds of bones or fresh bread.” That recipe started out: “Brown the salmon in oil, add creamed meat to the mixture.” It was making mistakes that showed the thing had no memory at all.
Spectrum: You say in the book that you can learn a lot about AI by giving it a task and watching it flail. What do you learn?
Shane: One thing you learn is how much it relies on surface appearances rather than deep understanding. With the recipes, for example: It got the structure of title, category, ingredients, instructions, yield at the end. But when you look more closely, it has instructions like “Fold the water and roll it into cubes.” So clearly this thing does not understand water, let alone the other things. It’s recognizing certain phrases that tend to occur, but it doesn’t have a concept that these recipes are describing something real. You start to realize how very narrow the algorithms in this world are. They only know exactly what we tell them in our data set.
BACK TO TOP↑ “The narrower the problem, the smarter the AI will seem” Spectrum: That makes me think of DeepMind’s AlphaGo, which was universally hailed as a triumph for AI. It can play the game of Go better than any human, but it doesn’t know what Go is. It doesn’t know that it’s playing a game.
Shane: It doesn’t know what a human is, or if it’s playing against a human or another program. That’s also a nice illustration of how well these algorithms do when they have a really narrow and well-defined problem.
The narrower the problem, the smarter the AI will seem. If it’s not just doing something repeatedly but instead has to understand something, coherence goes down. For example, take an algorithm that can generate images of objects. If the algorithm is restricted to birds, it could do a recognizable bird. If this same algorithm is asked to generate images of any animal, if its task is that broad, the bird it generates becomes an unrecognizable brown feathered smear against a green background.
Spectrum: That sounds… disturbing.
Shane: It’s disturbing in a weird amusing way. What’s really disturbing is the humans it generates. It hasn’t seen them enough times to have a good representation, so you end up with an amorphous, usually pale-faced thing with way too many orifices. If you asked it to generate an image of a person eating pizza, you’ll have blocks of pizza texture floating around. But if you give that image to an image-recognition algorithm that was trained on that same data set, it will say, “Oh yes, that’s a person eating pizza.”
BACK TO TOP↑ Why overestimating AI is dangerous Spectrum: Do you see it as your role to puncture the AI hype?
Shane: I do see it that way. Not a lot of people are bringing out this side of AI. When I first started posting my results, I’d get people saying, “I don’t understand, this is AI, shouldn’t it be better than this? Why doesn't it understand?” Many of the impressive examples of AI have a really narrow task, or they’ve been set up to hide how little understanding it has. There’s a motivation, especially among people selling products based on AI, to represent the AI as more competent and understanding than it actually is.
Spectrum: If people overestimate the abilities of AI, what risk does that pose?
Shane: I worry when I see people trusting AI with decisions it can’t handle, like hiring decisions or decisions about moderating content. These are really tough tasks for AI to do well on. There are going to be a lot of glitches. I see people saying, “The computer decided this so it must be unbiased, it must be objective.”

“If the algorithm’s task is to replicate human hiring decisions, it’s going to glom onto gender bias and race bias.”
—Janelle Shane, AI Weirdness blogger
That’s another thing I find myself highlighting in the work I’m doing. If the data includes bias, the algorithm will copy that bias. You can’t tell it not to be biased, because it doesn’t understand what bias is. I think that message is an important one for people to understand.
If there’s bias to be found, the algorithm is going to go after it. It’s like, “Thank goodness, finally a signal that’s reliable.” But for a tough problem like: Look at these resumes and decide who’s best for the job. If its task is to replicate human hiring decisions, it’s going to glom onto gender bias and race bias. There’s an example in the book of a hiring algorithm that Amazon was developing that discriminated against women, because the historical data it was trained on had that gender bias.
Spectrum: What are the other downsides of using AI systems that don’t really understand their tasks?
Shane: There is a risk in putting too much trust in AI and not examining its decisions. Another issue is that it can solve the wrong problems, without anyone realizing it. There have been a couple of cases in medicine. For example, there was an algorithm that was trained to recognize things like skin cancer. But instead of recognizing the actual skin condition, it latched onto signals like the markings a surgeon makes on the skin, or a ruler placed there for scale. It was treating those things as a sign of skin cancer. It’s another indication that these algorithms don’t understand what they’re looking at and what the goal really is.
BACK TO TOP↑ Giraffing Spectrum: In your blog, you often have neural nets generate names for things—such as ice cream flavors, paint colors, cats, mushrooms, and types of apples. How do you decide on topics?
Shane: Quite often it’s because someone has written in with an idea or a data set. They’ll say something like, “I’m the MIT librarian and I have a whole list of MIT thesis titles.” That one was delightful. Or they’ll say, “We are a high school robotics team, and we know where there’s a list of robotics team names.” It’s fun to peek into a different world. I have to be careful that I’m not making fun of the naming conventions in the field. But there’s a lot of humor simply in the neural net’s complete failure to understand. Puns in particular—it really struggles with puns.
Spectrum: Your blog is quite absurd, but it strikes me that machine learning is often absurd in itself. Can you explain the concept of giraffing?
Shane: This concept was originally introduced by [internet security expert] Melissa Elliott. She proposed this phrase as a way to describe the algorithms’ tendency to see giraffes way more often than would be likely in the real world. She posted a whole bunch of examples, like a photo of an empty field in which an image-recognition algorithm has confidently reported that there are giraffes. Why does it think giraffes are present so often when they’re actually really rare? Because they’re trained on data sets from online. People tend to say, “Hey look, a giraffe!” And then take a photo and share it. They don’t do that so often when they see an empty field with rocks.
There’s also a chatbot that has a delightful quirk. If you show it some photo and ask it how many giraffes are in the picture, it will always answer with some non zero number. This quirk comes from the way the training data was generated: These were questions asked and answered by humans online. People tended not to ask the question “How many giraffes are there?” when the answer was zero. So you can show it a picture of someone holding a Wii remote. If you ask it how many giraffes are in the picture, it will say two.
BACK TO TOP↑ Machine and human creativity Spectrum: AI can be absurd, and maybe also creative. But you make the point that AI art projects are really human-AI collaborations: Collecting the data set, training the algorithm, and curating the output are all artistic acts on the part of the human. Do you see your work as a human-AI art project?
Shane: Yes, I think there is artistic intent in my work; you could call it literary or visual. It’s not so interesting to just take a pre-trained algorithm that’s been trained on utilitarian data, and tell it to generate a bunch of stuff. Even if the algorithm isn’t one that I’ve trained myself, I think about, what is it doing that’s interesting, what kind of story can I tell around it, and what do I want to show people.

The Halloween costume algorithm “was able to draw on its knowledge of which words are related to suggest things like sexy barnacle.”
—Janelle Shane, AI Weirdness blogger
Spectrum: For the past three years you’ve been getting neural nets to generate ideas for Halloween costumes. As language models have gotten dramatically better over the past three years, are the costume suggestions getting less absurd?
Shane: Yes. Before I would get a lot more nonsense words. This time I got phrases that were related to real things in the data set. I don’t believe the training data had the words Flying Dutchman or barnacle. But it was able to draw on its knowledge of which words are related to suggest things like sexy barnacle and sexy Flying Dutchman.
Spectrum: This year, I saw on Twitter that someone made the gothy giraffe costume happen. Would you ever dress up for Halloween in a costume that the neural net suggested?
Shane: I think that would be fun. But there would be some challenges. I would love to go as the sexy Flying Dutchman. But my ambition may constrict me to do something more like a list of leg parts.
BACK TO TOP↑ Continue reading

Posted in Human Robots

#436186 Video Friday: Invasion of the Mini ...

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!):

DARPA SubT Urban Circuit – February 18-27, 2020 – Olympia, Wash., USA
Let us know if you have suggestions for next week, and enjoy today’s videos.

There will be a Mini-Cheetah Workshop (sponsored by Naver Labs) a year from now at IROS 2020 in Las Vegas. Mini-Cheetahs for everyone!

That’s just a rendering, of course, but this isn’t:

[ MCW ]

I was like 95 percent sure that the Urban Circuit of the DARPA SubT Challenge was going to be in something very subway station-y. Oops!

In the Subterranean (SubT) Challenge, teams deploy autonomous ground and aerial systems to attempt to map, identify, and report artifacts along competition courses in underground environments. The artifacts represent items a first responder or service member may encounter in unknown underground sites. This video provides a preview of the Urban Circuit event location. The Urban Circuit is scheduled for February 18-27, 2020, at Satsop Business Park west of Olympia, Washington.

[ SubT ]

Researchers at SEAS and the Wyss Institute for Biologically Inspired Engineering have developed a resilient RoboBee powered by soft artificial muscles that can crash into walls, fall onto the floor, and collide with other RoboBees without being damaged. It is the first microrobot powered by soft actuators to achieve controlled flight.

To solve the problem of power density, the researchers built upon the electrically-driven soft actuators developed in the lab of David Clarke, the Extended Tarr Family Professor of Materials. These soft actuators are made using dielectric elastomers, soft materials with good insulating properties, that deform when an electric field is applied. By improving the electrode conductivity, the researchers were able to operate the actuator at 500 Hertz, on par with the rigid actuators used previously in similar robots.

Next, the researchers aim to increase the efficiency of the soft-powered robot, which still lags far behind more traditional flying robots.

[ Harvard ]

We present a system for fast and robust handovers with a robot character, together with a user study investigating the effect of robot speed and reaction time on perceived interaction quality. The system can match and exceed human speeds and confirms that users prefer human-level timing.

In a 3×3 user study, we vary the speed of the robot and add variable sensorimotor delays. We evaluate the social perception of the robot using the Robot Social Attribute Scale (RoSAS). Inclusion of a small delay, mimicking the delay of the human sensorimotor system, leads to an improvement in perceived qualities over both no delay and long delay conditions. Specifically, with no delay the robot is perceived as more discomforting and with a long delay, it is perceived as less warm.

[ Disney Research ]

When cars are autonomous, they’re not going to be able to pump themselves full of gas. Or, more likely, electrons. Kuka has the solution.

[ Kuka ]

This looks like fun, right?

[ Robocoaster ]

NASA is leading the way in the use of On-orbit Servicing, Assembly, and Manufacturing to enable large, persistent, upgradable, and maintainable spacecraft. This video was developed by the Advanced Concepts Lab (ACL) at NASA Langley Research Center.

[ NASA ]

The noisiest workshop by far at Humanoids last month (by far) was Musical Interactions With Humanoids, the end result of which was this:

[ Workshop ]

IROS is an IEEE event, and in furthering the IEEE mission to benefit humanity through technological innovation, IROS is doing a great job. But don’t take it from us – we are joined by IEEE President-Elect Professor Toshio Fukuda to find out a bit more about the impact events like IROS can have, as well as examine some of the issues around intelligent robotics and systems – from privacy to transparency of the systems at play.

[ IROS ]

Speaking of IROS, we hope you’ve been enjoying our coverage. We have already featured Harvard’s strange sea-urchin-inspired robot and a Japanese quadruped that can climb vertical ladders, with more stories to come over the next several weeks.

In the mean time, enjoy these 10 videos from the conference (as usual, we’re including the title, authors, and abstract for each—if you’d like more details about any of these projects, let us know and we’ll find out more for you).

“A Passive Closing, Tendon Driven, Adaptive Robot Hand for Ultra-Fast, Aerial Grasping and Perching,” by Andrew McLaren, Zak Fitzgerald, Geng Gao, and Minas Liarokapis from the University of Auckland, New Zealand.

Current grasping methods for aerial vehicles are slow, inaccurate and they cannot adapt to any target object. Thus, they do not allow for on-the-fly, ultra-fast grasping. In this paper, we present a passive closing, adaptive robot hand design that offers ultra-fast, aerial grasping for a wide range of everyday objects. We investigate alternative uses of structural compliance for the development of simple, adaptive robot grippers and hands and we propose an appropriate quick release mechanism that facilitates an instantaneous grasping execution. The quick release mechanism is triggered by a simple distance sensor. The proposed hand utilizes only two actuators to control multiple degrees of freedom over three fingers and it retains the superior grasping capabilities of adaptive grasping mechanisms, even under significant object pose or other environmental uncertainties. The hand achieves a grasping time of 96 ms, a maximum grasping force of 56 N and it is able to secure objects of various shapes at high speeds. The proposed hand can serve as the end-effector of grasping capable Unmanned Aerial Vehicle (UAV) platforms and it can offer perching capabilities, facilitating autonomous docking.

“Unstructured Terrain Navigation and Topographic Mapping With a Low-Cost Mobile Cuboid Robot,” by Andrew S. Morgan, Robert L. Baines, Hayley McClintock, and Brian Scassellati from Yale University, USA.

Current robotic terrain mapping techniques require expensive sensor suites to construct an environmental representation. In this work, we present a cube-shaped robot that can roll through unstructured terrain and construct a detailed topographic map of the surface that it traverses in real time with low computational and monetary expense. Our approach devolves many of the complexities of locomotion and mapping to passive mechanical features. Namely, rolling movement is achieved by sequentially inflating latex bladders that are located on four sides of the robot to destabilize and tip it. Sensing is achieved via arrays of fine plastic pins that passively conform to the geometry of underlying terrain, retracting into the cube. We developed a topography by shade algorithm to process images of the displaced pins to reconstruct terrain contours and elevation. We experimentally validated the efficacy of the proposed robot through object mapping and terrain locomotion tasks.

“Toward a Ballbot for Physically Leading People: A Human-Centered Approach,” by Zhongyu Li and Ralph Hollis from Carnegie Mellon University, USA.

This work presents a new human-centered method for indoor service robots to provide people with physical assistance and active guidance while traveling through congested and narrow spaces. As most previous work is robot-centered, this paper develops an end-to-end framework which includes a feedback path of the measured human positions. The framework combines a planning algorithm and a human-robot interaction module to guide the led person to a specified planned position. The approach is deployed on a person-size dynamically stable mobile robot, the CMU ballbot. Trials were conducted where the ballbot physically led a blindfolded person to safely navigate in a cluttered environment.

“Achievement of Online Agile Manipulation Task for Aerial Transformable Multilink Robot,” by Fan Shi, Moju Zhao, Tomoki Anzai, Keita Ito, Xiangyu Chen, Kei Okada, and Masayuki Inaba from the University of Tokyo, Japan.

Transformable aerial robots are favorable in aerial manipulation tasks for their flexible ability to change configuration during the flight. By assuming robot keeping in the mild motion, the previous researches sacrifice aerial agility to simplify the complex non-linear system into a single rigid body with a linear controller. In this paper, we present a framework towards agile swing motion for the transformable multi-links aerial robot. We introduce a computational-efficient non-linear model predictive controller and joints motion primitive frame-work to achieve agile transforming motions and validate with a novel robot named HYRURS-X. Finally, we implement our framework under a table tennis task to validate the online and agile performance.

“Small-Scale Compliant Dual Arm With Tail for Winged Aerial Robots,” by Alejandro Suarez, Manuel Perez, Guillermo Heredia, and Anibal Ollero from the University of Seville, Spain.

Winged aerial robots represent an evolution of aerial manipulation robots, replacing the multirotor vehicles by fixed or flapping wing platforms. The development of this morphology is motivated in terms of efficiency, endurance and safety in some inspection operations where multirotor platforms may not be suitable. This paper presents a first prototype of compliant dual arm as preliminary step towards the realization of a winged aerial robot capable of perching and manipulating with the wings folded. The dual arm provides 6 DOF (degrees of freedom) for end effector positioning in a human-like kinematic configuration, with a reach of 25 cm (half-scale w.r.t. the human arm), and 0.2 kg weight. The prototype is built with micro metal gear motors, measuring the joint angles and the deflection with small potentiometers. The paper covers the design, electronics, modeling and control of the arms. Experimental results in test-bench validate the developed prototype and its functionalities, including joint position and torque control, bimanual grasping, the dynamic equilibrium with the tail, and the generation of 3D maps with laser sensors attached at the arms.

“A Novel Small-Scale Turtle-inspired Amphibious Spherical Robot,” by Huiming Xing, Shuxiang Guo, Liwei Shi, Xihuan Hou, Yu Liu, Huikang Liu, Yao Hu, Debin Xia, and Zan Li from Beijing Institute of Technology, China.

This paper describes a novel small-scale turtle-inspired Amphibious Spherical Robot (ASRobot) to accomplish exploration tasks in the restricted environment, such as amphibious areas and narrow underwater cave. A Legged, Multi-Vectored Water-Jet Composite Propulsion Mechanism (LMVWCPM) is designed with four legs, one of which contains three connecting rod parts, one water-jet thruster and three joints driven by digital servos. Using this mechanism, the robot is able to walk like amphibious turtles on various terrains and swim flexibly in submarine environment. A simplified kinematic model is established to analyze crawling gaits. With simulation of the crawling gait, the driving torques of different joints contributed to the choice of servos and the size of links of legs. Then we also modeled the robot in water and proposed several underwater locomotion. In order to assess the performance of the proposed robot, a series of experiments were carried out in the lab pool and on flat ground using the prototype robot. Experiments results verified the effectiveness of LMVWCPM and the amphibious control approaches.

“Advanced Autonomy on a Low-Cost Educational Drone Platform,” by Luke Eller, Theo Guerin, Baichuan Huang, Garrett Warren, Sophie Yang, Josh Roy, and Stefanie Tellex from Brown University, USA.

PiDrone is a quadrotor platform created to accompany an introductory robotics course. Students build an autonomous flying robot from scratch and learn to program it through assignments and projects. Existing educational robots do not have significant autonomous capabilities, such as high-level planning and mapping. We present a hardware and software framework for an autonomous aerial robot, in which all software for autonomy can run onboard the drone, implemented in Python. We present an Unscented Kalman Filter (UKF) for accurate state estimation. Next, we present an implementation of Monte Carlo (MC) Localization and Fast-SLAM for Simultaneous Localization and Mapping (SLAM). The performance of UKF, localization, and SLAM is tested and compared to ground truth, provided by a motion-capture system. Our evaluation demonstrates that our autonomous educational framework runs quickly and accurately on a Raspberry Pi in Python, making it ideal for use in educational settings.

“FlightGoggles: Photorealistic Sensor Simulation for Perception-driven Robotics using Photogrammetry and Virtual Reality,” by Winter Guerra, Ezra Tal, Varun Murali, Gilhyun Ryou and Sertac Karaman from the Massachusetts Institute of Technology, USA.

FlightGoggles is a photorealistic sensor simulator for perception-driven robotic vehicles. The key contributions of FlightGoggles are twofold. First, FlightGoggles provides photorealistic exteroceptive sensor simulation using graphics assets generated with photogrammetry. Second, it provides the ability to combine (i) synthetic exteroceptive measurements generated in silico in real time and (ii) vehicle dynamics and proprioceptive measurements generated in motio by vehicle(s) in flight in a motion-capture facility. FlightGoggles is capable of simulating a virtual-reality environment around autonomous vehicle(s) in flight. While a vehicle is in flight in the FlightGoggles virtual reality environment, exteroceptive sensors are rendered synthetically in real time while all complex dynamics are generated organically through natural interactions of the vehicle. The FlightGoggles framework allows for researchers to accelerate development by circumventing the need to estimate complex and hard-to-model interactions such as aerodynamics, motor mechanics, battery electrochemistry, and behavior of other agents. The ability to perform vehicle-in-the-loop experiments with photorealistic exteroceptive sensor simulation facilitates novel research directions involving, e.g., fast and agile autonomous flight in obstacle-rich environments, safe human interaction, and flexible sensor selection. FlightGoggles has been utilized as the main test for selecting nine teams that will advance in the AlphaPilot autonomous drone racing challenge. We survey approaches and results from the top AlphaPilot teams, which may be of independent interest. FlightGoggles is distributed as open-source software along with the photorealistic graphics assets for several simulation environments, under the MIT license at http://flightgoggles.mit.edu.

“An Autonomous Quadrotor System for Robust High-Speed Flight Through Cluttered Environments Without GPS,” by Marc Rigter, Benjamin Morrell, Robert G. Reid, Gene B. Merewether, Theodore Tzanetos, Vinay Rajur, KC Wong, and Larry H. Matthies from University of Sydney, Australia; NASA Jet Propulsion Laboratory, California Institute of Technology, USA; and Georgia Institute of Technology, USA.

Robust autonomous flight without GPS is key to many emerging drone applications, such as delivery, search and rescue, and warehouse inspection. These and other appli- cations require accurate trajectory tracking through cluttered static environments, where GPS can be unreliable, while high- speed, agile, flight can increase efficiency. We describe the hardware and software of a quadrotor system that meets these requirements with onboard processing: a custom 300 mm wide quadrotor that uses two wide-field-of-view cameras for visual- inertial motion tracking and relocalization to a prior map. Collision-free trajectories are planned offline and tracked online with a custom tracking controller. This controller includes compensation for drag and variability in propeller performance, enabling accurate trajectory tracking, even at high speeds where aerodynamic effects are significant. We describe a system identification approach that identifies quadrotor-specific parameters via maximum likelihood estimation from flight data. Results from flight experiments are presented, which 1) validate the system identification method, 2) show that our controller with aerodynamic compensation reduces tracking error by more than 50% in both horizontal flights at up to 8.5 m/s and vertical flights at up to 3.1 m/s compared to the state-of-the-art, and 3) demonstrate our system tracking complex, aggressive, trajectories.

“Morphing Structure for Changing Hydrodynamic Characteristics of a Soft Underwater Walking Robot,” by Michael Ishida, Dylan Drotman, Benjamin Shih, Mark Hermes, Mitul Luhar, and Michael T. Tolley from the University of California, San Diego (UCSD) and University of Southern California, USA.

Existing platforms for underwater exploration and inspection are often limited to traversing open water and must expend large amounts of energy to maintain a position in flow for long periods of time. Many benthic animals overcome these limitations using legged locomotion and have different hydrodynamic profiles dictated by different body morphologies. This work presents an underwater legged robot with soft legs and a soft inflatable morphing body that can change shape to influence its hydrodynamic characteristics. Flow over the morphing body separates behind the trailing edge of the inflated shape, so whether the protrusion is at the front, center, or back of the robot influences the amount of drag and lift. When the legged robot (2.87 N underwater weight) needs to remain stationary in flow, an asymmetrically inflated body resists sliding by reducing lift on the body by 40% (from 0.52 N to 0.31 N) at the highest flow rate tested while only increasing drag by 5.5% (from 1.75 N to 1.85 N). When the legged robot needs to walk with flow, a large inflated body is pushed along by the flow, causing the robot to walk 16% faster than it would with an uninflated body. The body shape significantly affects the ability of the robot to walk against flow as it is able to walk against 0.09 m/s flow with the uninflated body, but is pushed backwards with a large inflated body. We demonstrate that the robot can detect changes in flow velocity with a commercial force sensor and respond by morphing into a hydrodynamically preferable shape. Continue reading

Posted in Human Robots

#436180 Bipedal Robot Cassie Cal Learns to ...

There’s no particular reason why knowing how to juggle would be a useful skill for a robot. Despite this, robots are frequently taught how to juggle things. Blind robots can juggle, humanoid robots can juggle, and even drones can juggle. Why? Because juggling is hard, man! You have to think about a bunch of different things at once, and also do a bunch of different things at once, which this particular human at least finds to be overly stressful. While juggling may not stress robots out, it does require carefully coordinated sensing and computing and actuation, which means that it’s as good a task as any (and a more entertaining task than most) for testing the capabilities of your system.

UC Berkeley’s Cassie Cal robot, which consists of two legs and what could be called a torso if you were feeling charitable, has just learned to juggle by bouncing a ball on what would be her head if she had one of those. The idea is that if Cassie can juggle while balancing at the same time, she’ll be better able to do other things that require dynamic multitasking, too. And if that doesn’t work out, she’ll still be able to join the circus.

Cassie’s juggling is assisted by an external motion capture system that tracks the location of the ball, but otherwise everything is autonomous. Cassie is able to juggle the ball by leaning forwards and backwards, left and right, and moving up and down. She does this while maintaining her own balance, which is the whole point of this research—successfully executing two dynamic behaviors that may sometimes be at odds with one another. The end goal here is not to make a better juggling robot, but rather to explore dynamic multitasking, a skill that robots will need in order to be successful in human environments.

This work is from the Hybrid Robotics Lab at UC Berkeley, led by Koushil Sreenath, and is being done by Katherine Poggensee, Albert Li, Daniel Sotsaikich, Bike Zhang, and Prasanth Kotaru.

For a bit more detail, we spoke with Albert Li via email.

Image: UC Berkeley

UC Berkeley’s Cassie Cal getting ready to juggle.

IEEE Spectrum: What would be involved in getting Cassie to juggle without relying on motion capture?

Albert Li: Our motivation for starting off with motion capture was to first address the control challenge of juggling on a biped without worrying about implementing the perception. We actually do have a ball detector working on a camera, which would mean we wouldn’t have to rely on the motion capture system. However, we need to mount the camera in a way that it would provide the best upwards field of view, and we also have develop a reliable estimator. The estimator is particularly important because when the ball gets close enough to the camera, we actually can’t track the ball and have to assume our dynamic models describe its motion accurately enough until it bounces back up.

What keeps Cassie from juggling indefinitely?

There are a few factors that affect how long Cassie can sustain a juggle. While in simulation the paddle exhibits homogeneous properties like its stiffness and damping, in reality every surface has anisotropic contact properties. So, there are parts of the paddle which may be better for juggling than others (and importantly, react differently than modeled). These differences in contact are also exacerbated due to how the paddle is cantilevered when mounted on Cassie. When the ball hits these areas, it leads to a larger than expected error in a juggle. Due to the small size of the paddle, the ball may then just hit the paddle’s edge and end the juggling run. Over a very long run, this is a likely occurrence. Additionally, some large juggling errors could cause Cassie’s feet to slip slightly, which ends up changing the stable standing position over time. Since this version of the controller assumes Cassie is stationary, this change in position eventually leads to poor juggles and failure.

Would Cassie be able to juggle while walking (or hovershoe-ing)?

Walking (and hovershoe-ing) while juggling is a far more challenging problem and is certainly a goal for future research. Some of these challenges include getting the paddle to precise poses to juggle the ball while also moving to avoid any destabilizing effects of stepping incorrectly. The number of juggles per step of walking could also vary and make the mathematics of the problem more challenging. The controller goal is also more involved. While the current goal of the juggling controller is to juggle the ball to a static apex position, with a walking juggling controller, we may instead want to hit the ball forwards and also walk forwards to bounce it, juggle the ball along a particular path, etc. Solving such challenges would be the main thrusts of the follow-up research.

Can you give an example of a practical task that would be made possible by using a controller like this?

Studying juggling means studying contact behavior and leveraging our models of it to achieve a known objective. Juggling could also be used to study predictable post-contact flight behavior. Consider the scenario where a robot is attempting to make a catch, but fails, letting the ball to bounce off of its hand, and then recovering the catch. This behavior could also be intentional: It is often easier to first execute a bounce to direct the target and then perform a subsequent action. For example, volleyball players could in principle directly hit a spiked ball back, but almost always bump the ball back up and then return it.

Even beyond this motivating example, the kinds of models we employ to get juggling working are more generally applicable to any task that involves contact, which could include tasks besides bouncing like sliding and rolling. For example, clearing space on a desk by pushing objects to the side may be preferable than individually manipulating each and every object on it.

You mention collaborative juggling or juggling multiple balls—is that something you’ve tried yet? Can you talk a bit more about what you’re working on next?

We haven’t yet started working on collaborative or multi-ball juggling, but that’s also a goal for future work. Juggling multiple balls statically is probably the most reasonable next goal, but presents additional challenges. For instance, you have to encode a notion of juggling urgency (if the second ball isn’t hit hard enough, you have less time to get the first ball up before you get back to the second one).

On the other hand, collaborative human-robot juggling requires a more advanced decision-making framework. To get robust multi-agent juggling, the robot will need to employ some sort of probabilistic model of the expected human behavior (are they likely to move somewhere? Are they trying to catch the ball high or low? Is it safe to hit the ball back?). In general, developing such human models is difficult since humans are fairly unpredictable and often don’t exhibit rational behavior. This will be a focus of future work.

[ Hybrid Robotics Lab ] Continue reading

Posted in Human Robots

#436176 We’re Making Progress in Explainable ...

Machine learning algorithms are starting to exceed human performance in many narrow and specific domains, such as image recognition and certain types of medical diagnoses. They’re also rapidly improving in more complex domains such as generating eerily human-like text. We increasingly rely on machine learning algorithms to make decisions on a wide range of topics, from what we collectively spend billions of hours watching to who gets the job.

But machine learning algorithms cannot explain the decisions they make.
How can we justify putting these systems in charge of decisions that affect people’s lives if we don’t understand how they’re arriving at those decisions?

This desire to get more than raw numbers from machine learning algorithms has led to a renewed focus on explainable AI: algorithms that can make a decision or take an action, and tell you the reasons behind it.

What Makes You Say That?
In some circumstances, you can see a road to explainable AI already. Take OpenAI’s GTP-2 model, or IBM’s Project Debater. Both of these generate text based on a large corpus of training data, and try to make it as relevant as possible to the prompt that’s given. If these models were also able to provide a quick run-down of the top few sources in that corpus of training data they were drawing information from, it may be easier to understand where the “argument” (or poetic essay about unicorns) was coming from.

This is similar to the approach Google is now looking at for its image classifiers. Many algorithms are more sensitive to textures and the relationship between adjacent pixels in an image, rather than recognizing objects by their outlines as humans do. This leads to strange results: some algorithms can happily identify a totally scrambled image of a polar bear, but not a polar bear silhouette.

Previous attempts to make image classifiers explainable relied on significance mapping. In this method, the algorithm would highlight the areas of the image that contributed the most statistical weight to making the decision. This is usually determined by changing groups of pixels in the image and seeing which contribute to the biggest change in the algorithm’s impression of what the image is. For example, if the algorithm is trying to recognize a stop sign, changing the background is unlikely to be as important as changing the sign.

Google’s new approach changes the way that its algorithm recognizes objects, by examining them at several different resolutions and searching for matches to different “sub-objects” within the main object. You or I might recognize an ambulance from its flashing lights, its tires, and its logo; we might zoom in on the basketball held by an NBA player to deduce their occupation, and so on. By linking the overall categorization of an image to these “concepts,” the algorithm can explain its decision: I categorized this as a cat because of its tail and whiskers.

Even in this experiment, though, the “psychology” of the algorithm in decision-making is counter-intuitive. For example, in the basketball case, the most important factor in making the decision was actually the player’s jerseys rather than the basketball.

Can You Explain What You Don’t Understand?
While it may seem trivial, the conflict here is a fundamental one in approaches to artificial intelligence. Namely, how far can you get with mere statistical associations between huge sets of data, and how much do you need to introduce abstract concepts for real intelligence to arise?

At one end of the spectrum, Good Old-Fashioned AI or GOFAI dreamed up machines that would be entirely based on symbolic logic. The machine would be hard-coded with the concept of a dog, a flower, cars, and so forth, alongside all of the symbolic “rules” which we internalize, allowing us to distinguish between dogs, flowers, and cars. (You can imagine a similar approach to a conversational AI would teach it words and strict grammatical structures from the top down, rather than “learning” languages from statistical associations between letters and words in training data, as GPT-2 broadly does.)

Such a system would be able to explain itself, because it would deal in high-level, human-understandable concepts. The equation is closer to: “ball” + “stitches” + “white” = “baseball”, rather than a set of millions of numbers linking various pathways together. There are elements of GOFAI in Google’s new approach to explaining its image recognition: the new algorithm can recognize objects based on the sub-objects they contain. To do this, it requires at least a rudimentary understanding of what those sub-objects look like, and the rules that link objects to sub-objects, such as “cats have whiskers.”

The issue, of course, is the—maybe impossible—labor-intensive task of defining all these symbolic concepts and every conceivable rule that could possibly link them together by hand. The difficulty of creating systems like this, which could handle the “combinatorial explosion” present in reality, helped to lead to the first AI winter.

Meanwhile, neural networks rely on training themselves on vast sets of data. Without the “labeling” of supervised learning, this process might bear no relation to any concepts a human could understand (and therefore be utterly inexplicable).

Somewhere between these two, hope explainable AI enthusiasts, is a happy medium that can crunch colossal amounts of data, giving us all of the benefits that recent, neural-network AI has bestowed, while showing its working in terms that humans can understand.

Image Credit: Image by Seanbatty from Pixabay Continue reading

Posted in Human Robots