#toyota

#toyota

Researchers at Toyota Have Successfully Taught a Car to Autonomously DriftProfessional drivers could probably handle extreme road conditions and come out unscathed in the end. For non-professional drivers, however, that statement probably won't be true. They don't have the skills of a pro, after all. It is no secret that numerous car accidents happen every year, and this makes us ask: how do we prevent car accidents even in extreme situations? The Toyota Research Institute is currently looking for answers to this question, and they have already developed a system that could be a potential solution to preventing said accidents.In this video, the Toyota Research Institute showcases a prototype vehicle equipped with a new active safety system that enables the car to drift around obstacles autonomously — a technology that temporarily amplifies the skills of a regular driver. The system also can calculate a whole new trajectory every twentieth of a second to maintain the car's balance as it navigates.We need this technology installed in every car!Image Credit: Toyota Research Institute/ YouTube#Car #Innovation #Safety #Drifting #Toyota #ToyotaResearchInstitute

Toyota's New Robot Can Clean Home Surfaces While Taking a Robo-SelfieIf Toyota succeeds, perhaps one day soon its robot could clean your house so you won't have to!Toyota Research Institute (TRI) announced that its new robot can understand and operate in complicated environments (like a house) that confuse most other robots, such as recognizing transparent and reflective surfaces.MAx Bajracharya, vice president of robotics at TRI explains:While a human can easily differentiate between an object and its reflection, transparent or reflective items commonly found in the home befuddle today’s robots. Since most robots are programmed to react to the objects and geometry in front of them without considering the context of the situation, they are easily fooled by a glass table, shiny toaster or transparent cup.“To overcome this, TRI roboticists developed a novel training method to perceive the 3D geometry of the scene while also detecting objects and surfaces,” continued Bajracharya. “This combination enables researchers to use large amounts of synthetic data to train the system.” Using synthetic data also alleviates the need for time consuming, expensive, or impractical data collection and labeling.#Toyota #robot #cleaning