Elon Musk | New robots with robot arms | High-Tech News
Engineers create Astro, a talented robot dog that can be trained with built-in artificial intelligence, deep learning and multiple sensors.
Florida Atlantic University team presented a robot that combines the advantages of voice assistants and dexterous Boston Dynamic devices. Astro does not use pre-programmed automation algorithms, but learns new things using a neural network that analyzes its own experience and incoming information, by analogy with the brain of real dogs.
The robot weighing 45 kg is equipped with optical, sound, gas sensors, high-tech radar, cameras and directional training microphone. Like a regular pet, it performs commands such as «sit», «stand» and «lie down». Over time, he will be able to understand and respond to hand movements, distinguish colors, know many languages, interact with drones, distinguish human faces, and even recognize other dogs..
According to the developers, Astro will be able to help the police, military and security services in the search for weapons, explosives and the remnants of shells. It can be programmed to help the visually impaired or to carry out medical diagnostics. The team is also training a robot dog for first aid in search and rescue missions..
Mechanical Doberman pinscher interacts and reacts with the outside world in real time, will be able to move over rough terrain, protect people and animals in the event of dangerous situations, quickly see and distinguish thousands of faces in the database, smell the air to detect foreign substances, and also hear and respond to requests for help that are out of the range of human hearing.
To process incoming sensory data and make independent behavioral decisions, Astro has a suite of Nvidia Jetson TX2 GPUs with a combined processing power of four teraflops, which is about four trillion calculations per second..
The development of robots allows them to be used not only for demonstrations, but also in everyday life. So recently robotic patrol started helping the traffic police of China
text: Ilya Bauer, photo and video: Florida Atlantic University