Human Robot Interaction

Artificial cognition for social human–robot interaction: An implementation


We identify first the needed individual and collaborative cognitive skills: geometric reasoning and situation assessment based on perspective-taking and affordance analysis; acquisition and representation of knowledge models for multiple agents (humans and robots, with their specificities); situated, natural and multi-modal dialogue; human-aware task planning; human–robot joint task achievement. It lays at the crossroad of many subdomains of AI and, in effect, it calls for their integration: modelling humans and human cognition; acquiring, representing, manipulating in a tractable way abstract knowledge at the human level; reasoning on this knowledge to make decisions; eventually instantiating those decisions into physical actions both legible to and in coordination with humans. Figure 1 illustrates this context: the human and the robot share a common space and exchange information through multiple modalities (we specifically consider verbal communication, deictic gestures and social gaze), and the robot is expected to achieve interactive object manipulation, fetch and carry tasks and other similar chores by taking into account, at every stage, the intentions, beliefs, perspectives, skills of its human partner. Reasoning and planning take place at symbolic as well as geometric level and take into account agents beliefs, perspectives and capabilities (D) as estimated by the robot.

Autonomous cars with no human backup to hit the road

Daily Mail

Automotive electronics and parts maker Delphi and French transport company Transdev plan to use autonomous taxis and a shuttle van to carry passengers on roadways in France. Volvo launched'Drive Me UK' earlier this year, an extensive UK-based autonomous driving trial, involving up to 100 driverless cars being driven on roads by people later this year. But humans at a central dispatch center would still be able to take control of the vehicles, said Glen De Vos, Delphi Corp.'s chief technology officer. Bryant Walker Smith, a law professor at the University of South Carolina who tracks self-driving cars, said the Delphi-Transdev project would be among the first to carry passengers on roads without human backup drivers.

Less self-assured AI are unlikely to override human orders

Daily Mail

In the Terminator film franchise, hyper-intelligent robots learn to operate without their human masters, leading to a machine uprising that wipes out most of mankind. To test their theory on'robot-confidence', the researchers developed an algorithmic model of an interaction between humans and robots called the'off-switch game'. To test their theory on'robot-confidence', the researchers developed an algorithmic model of an interaction between humans and robots called the'off-switch game'. In the Terminator film franchise (pictured) hyper-intelligent robots learn to operate without their human masters.

Japan to launch self-navigating cargo ships 'by 2025' - BBC News


Japanese shipping companies are working with shipbuilders to develop self-piloting cargo ships. Shipping firms Mitsui OSK Lines and Nippon Yusen are working with shipbuilders including Japan Marine United to share both costs and expertise, according to the Nikkei Asian Review. Nippon Yusen has already been working on technology to enable ships to use data to assess collision risks. In 2016, Rolls-Royce announced plans to develop unmanned cargo ships, starting with remote-controlled vessels that could be operational as soon as 2020.

Program Allows Kids To Build And Fly Drones, While Helping Them Grow In STEM Fields

International Business Times

A new program called Project Icarus is teaching kids how to build drones, while at the same time helping them grow in STEM fields, which focus in science, technology, engineering and mathematics. Seven-year-old Ava builds her drone at a Project Icarus workshop. Dani Dias' seven-year-old daughter Ava participated in a Project Icarus workshop. The drone program helped Ava grow, her mother said.

Japan to operate self-navigating cargo ships soon


The plan to have so-called "smart ships" moving around the world is being driven by a consortium of Japanese shipping companies who are working with shipbuilders to develop self-piloting cargo ships. At the heart of the ships will be a form of artificial intelligence. The idea is to reduce costs and to improve efficiency. From then on a human captain will be based on-shore, monitoring the progress of the boats as they navigate the world's marine trade routes.

AI Will Help Self-Navigating Ships Plot Their Ways, First Fleet Expected By 2025

International Business Times

The AI tech is expected to make shipping safer by not just potting the course, but also detecting machinery malfunctions and other problems way in advance. Since Japan has a head start on the technology, it could make the country's shipping technology more competitive. The marine branch of carmaker Rolls Royce is also working on an autonomous shipping technology. Ever since then, many tech companies ride-hailing services and auto companies have ventured into the technology.

Self-driving car angels protect you from yourself WIRED


The driver keeps pressing the accelerator. But a bunch of electronics hacked into the car brings it to a safe stop anyway--the system had already been tracking the pedestrian for some time using lasers and a camera. The vehicle won't budge until the pedestrian is safely out of the way.

Video Friday: Extra Robot Arms, Anti-Drone Drone, and Adorable TurtleBots

IEEE Spectrum Robotics Channel

The Reconfigurable Robotics Lab presents Mori, a modular origami robot. Mori is the first example of a robot that combines the concepts behind both origami robots and reconfigurable, modular robots. Origami robotics utilises folding of thin structures to produce single robots that can change their shape, while modular robotics uses large numbers of individual entities to reconfigure the overall shape and address diverse tasks. UT's Human Centered Robotics Lab has been running some painful-looking simulations on Valkyrie: We'd all like to see this tried on the actual robot, I think.

Bosch plans to use radar sensors in millions of cars to make better maps

Popular Science

Self-driving cars need accurate maps. The maps should ideally be accurate down to less than an inch, says Christoph Mertz, a scientist at the Robotics Institute of Carnegie Mellon University and cofounder of a road monitoring company called Roadbotics. Maps of the future may get an accuracy boost, thanks to a new partnership between Bosch and mapping company TomTom; they are collaborating on a technology they call a "radar road signature." The idea, announced on Wednesday, is that cars driven by humans (and possibly autonomous ones as well) will use onboard radar sensors to map the roads in a highly-detailed way.