Autonomous car futurist Brad Templeton talks about the technical factors behind the fatal crash and discusses implications for the autonomous car industry.
New research from the Computational Neuroscience Lab at the University of Tartu looks at what happens when multiple AI agents are competing or collaborating in the same environment. The research (built on some of the pioneering Deep Reinforcement Learning work done by Deepmind) shows that AI agents will tend to try to maximize their rewards. Leaves interesting questions about the role of researchers in defining these rewards clearly and ethically.
California's new state driving laws effectively make it illegal to operate a vehicle without a licensed driver, making Google's plans for unmanned robocars (at least for now) impossible to pursue in the state, and giving an advantage to traditional car manufacturers, which are mostly pursuing co-pilot or auto-pilot versions of autonomous driving technology. Will this push robocar development to more autonomous-friendly states? Interesting article by robocar expert Brad Templeton.
The Robot Report has been compiling and vetting companies that manufacture robots or are directly involved in their making since 2008. Today that list has exceeded 4,000 manufacturers, integrators, service robot providers, startup companies, ancillary businesses and educational facilities. The list is global and free.
NTT’s recent entry into consumer robotics suggests that it’s not just about Google vs Amazon, or Softbank vs NTT anymore, but about telcos vs big data companies. With two major and distinct sectors now competing for market share in consumer robotics, will this impact the way the whole robotics industry evolves?
The release last week of the study on adverse events in robotic surgery led to much discussion on the safety and effectiveness of robotic surgical procedures. MIT Sloane's Matt Beane argues that while the hope is that this dialogue will mean safer and more effective robotic procedures in the future, the intense focus on safety and effectiveness has compromised training opportunities for new robotic surgeons, who require many hours of “live” surgical practice time to develop their skills. Beane says that robotic surgery will likely continue to expand in proportion to other methods, given that it allows fewer surgeons to perform surgery with less trauma to the patient, but no matter how safe we make robotic surgical procedures, they will become a luxury available to a very few if we fail to address the sustainability of the practice.
Back Sproutel’s crowdfunding campaign to get Jerry the Bear, their robotic healthcare coach, into the hands of every kid in the US who is newly diagnosed with type 1 diabetes.
Amazon just announced it will be launching a new drone delivery service called PrimeAir as early as 2015, but safety concerns will be key — no one wants an Amazon delivery falling out of the sky. As Amazon CEO Jeff Bezos says, "Look, this thing can’t land on somebody’s head while they’re walking around their neighborhood." Check out this new video by Mark Mueller out of Raffaello D'Andrea's Flying Machine Arena at ETH Zurich (note: D'Andrea is the tech wizard behind Amazon's robotic warehouse) — it shows a new failsafe algorithm that allows an unmanned aerial vehicle to recover and land safely following propeller loss. According to the researchers, the algorithm allows the vehicle to remain in flight despite the loss of one, two, or possibly even three propellers.
This project uses a unique design approach called 'adaptive morphology', where one part of a robot’s structure is adaptable and shared between different modes of operation. Cool!