{"id":14022,"date":"2024-06-25T09:45:41","date_gmt":"2024-06-25T01:45:41","guid":{"rendered":"https:\/\/www.1ai.net\/?p=14022"},"modified":"2024-06-25T09:45:41","modified_gmt":"2024-06-25T01:45:41","slug":"%e6%97%a5%e6%9c%ac%e5%bc%80%e5%8f%91%e4%ba%ba%e5%bd%a2%e6%9c%ba%e5%99%a8%e4%ba%baalter3%ef%bc%9a%e9%87%87%e7%94%a8gpt-4%e6%8a%80%e6%9c%af-%e8%87%aa%e6%8b%8d%e9%83%bd%e5%ad%a6%e4%bc%9a","status":"publish","type":"post","link":"https:\/\/www.1ai.net\/en\/14022.html","title":{"rendered":"Japan develops humanoid robot Alter3: It can even take selfies using GPT-4 technology"},"content":{"rendered":"

Japan<\/a>Researchers at the University of Tokyo, in collaboration with Alternative Machine, have made a breakthrough by developing a robot that can directly map natural language commands to robot actions.Humanoid Robot<\/a>System Alter3. Its background model adopts GPT-4<\/a>Technology that can complete a range of complex tasks, such as taking selfies or playing ghost.<\/p>\n

\"\"<\/p>\n

It\u2019s one of a growing number of studies that combine basic models with robotic systems. While these systems haven\u2019t yet reached scalable commercial solutions, they have advanced robotics research in recent years and show great potential.<\/p>\n

Alter3 uses GPT-4 technology as a back-office model to receive natural language instructions describing situations in which action or robots are required to react. First, the model uses the \"agent framework\" to plan a series of operational steps required for robots to achieve their goals. Second, through coding agents, robots are created to perform the commands required for each step. Since GPT-4 is not trained in Alter3 programming commands, researchers use their context learning skills to adapt their behaviour to robotic API\u3002<\/p>\n

Therefore, the prompt contains a list of commands and a set of examples of how to use each command. The model then maps each step to one or more API commands to send to the robot for execution.<\/p>\n

The researchers added functionality so that humans can provide feedback, such as "raise your arm a little higher." These instructions are sent to another GPT-4 agent, which reasons about the code, makes necessary corrections, and returns the action sequence to the robot. The improved action recipes and code are stored in a database for future use.<\/p>\n

The researchers ran several tests on Alter3, including everyday actions, such as taking selfies and drinking tea, and imitating actions, such as playing a ghost or a snake. They also tested the model's ability to cope with situations that require carefully planned actions. GPT-4's extensive understanding of human behavior and actions allows for the creation of more realistic behavior plans for humanoid robots such as Alter3. The researchers' experiments also showed that they were able to imitate emotions such as shame and joy in the robot.<\/p>","protected":false},"excerpt":{"rendered":"

Researchers at the University of Tokyo, in collaboration with Alternative Machine, have broken new ground by developing Alter3, a humanoid robot system that maps natural language commands directly to robotic actions, and whose backend model utilizes GPT-4 technology to perform a wide range of complex tasks, such as taking selfies or playing the role of a ghost. This is one of a growing number of research results based on combining basic models with robotic systems. While these systems have yet to reach a scalable commercial solution, they have pushed the boundaries of robotics research in recent years and show great potential. Alter3 uses GPT-4 technology as a backend model that receives natural language commands describing actions or scenarios that the robot needs to react to. First, the model<\/p>","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[146],"tags":[510,907,226],"collection":[],"class_list":["post-14022","post","type-post","status-publish","format-standard","hentry","category-news","tag-gpt-4","tag-907","tag-226"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/14022","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/comments?post=14022"}],"version-history":[{"count":0,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/14022\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/media?parent=14022"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/categories?post=14022"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/tags?post=14022"},{"taxonomy":"collection","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/collection?post=14022"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}