In just a few decades of development, robots have adapted to and can execute complex and diverse instructions. Now, empowered by AI, they are developing towards more advanced humanoid robot forms, perhaps in the near future. , will also be responsible for the continuation of carbon-based life.
文 | Ji Sheng
Source | "Manager" Magazine
On August 16, 2023, Beijing successfully hosted the World Robot Conference. There are more than 140 companies participating in this exhibition, and nearly 600 robot products are on display. Among them, there are more than 10 humanoid robot companies participating in the exhibition. Compared with only one company in previous years, the growth rate can be described as geometric
In addition, just before the exhibition, Beijing issued the "Several Measures to Promote the Innovation and Development of the Robot Industry" policy, including the establishment of a 10 billion yuan robot industry fund, and also mentioned the establishment of a humanoid robot innovation center. Does this mean that humanoid robots will soon What about ushering in the moment of development when "a hundred flowers bloom and a hundred schools of thought contend"?
In the production activities of human society, repetitive labor or dangerous jobs often occur, such as assembly lines in factories, daily transportation, security monitoring and patrols, dangerous goods production and inspection, etc. So is there a tool that can How about helping people get rid of these repetitive, mechanical, tedious and dangerous jobs, and thereby liberate productivity? In this context, robots that replace humans in completing some daily instructions emerged at the historic moment, and have developed rapidly under the stimulation of huge social needs.
Since the creation and use of the first industrial robot in the 1960s, the development history of the robot industry has been less than a hundred years, which is insignificant compared with the thousands of years of human civilization. However, in these short decades, robots have The industry has undergone earth-shaking changes, and there are countless types of corresponding robots. For example, industrial robots that meet the needs of manufacturing, service robots that serve daily life, special robots that deal with special scenes, and bionic robots that draw on perfect life systems...
From an economic point of view, demand can be created. Robots serve as a “jack-of-all-trades” tool that can help meet demand faster, and the size of the robotics market is difficult to estimate. In addition, humanoid robots, as a special type of bionic robot, will push various types of robots to eventually return to the humanoid foundation in today's social environment where humans are the basic unit, which will lay the core trend for the future development of robots
At this stage, there is no official unified definition and relevant standards for the description of humanoid robots on the market. In the author's opinion, the key to humanoid robots is "people". To become the incarnation of human beings, in addition to completing tasks that humans can basically complete In addition, it must do things that humans cannot currently do. To fulfill the above requirements, humanoid robots must have a sufficiently intelligent and flexible "brain body" so that they can better match the execution of "humans" standard.
The control core of the humanoid robot is called the "brain", which is responsible for processing signals such as the robot's perception, motion control, scheduling and coordination. This is the core criterion for measuring the intelligence of humanoid robots. In the trend of rapid development of artificial intelligence in 2023, more and more companies are applying large AI models to humanoid robots, making it an important technology transformation scenario for embodied intelligence. Although this technology is currently being questioned, it plays an important role in solving the logical understanding, visual perception and movement coordination abilities of humanoid robots. With the strengthening of basic AI large model simulation training, humanoid robots will be able to achieve more advanced high-dimensional perception of movement in 3D space, and are not limited to basic walking and simple task execution functions
With the support of a human-like "brain", the rapid implementation of humanoid robots also requires a strong "body", that is, the hardware level of humanoid robots must also be solved. The first is to solve the dynamic balance, which will put forward higher requirements for the motion control system of humanoid robots to realize the perception and detection of the surrounding environment, whole-body motion planning, joint motion control, overall coordination and scheduling, etc. At this stage, the motion control system mainly consists of motion controllers, servo systems (drivers and motors), feedback sensors, and numerous actuators. The initially developed industrial robots have a strong dependence on the use of this system, so that in this system The technical foundation in the field is relatively sufficient, and the subsequent development trend is gradually turning to the integration of drive and control, achieving higher efficiency by increasing flexibility, which also makes the development of humanoid robots take fewer detours
Secondly, with the support of the brain and flexible mobility, efforts are made to create the five senses of humanoid robots. Currently, one of the five senses of humanoid robots, vision, mainly uses 3D vision sensors, lidar, etc. The technology in this area is not yet mature, and it is not yet mature. Continuous iterative upgrades are required; the haptic mainly uses force sensor solutions. According to data released by Blue Point Touch, the six-dimensional force sensor at the end of the humanoid robot has a range of 1000N, and the force measurement accuracy is within five thousandths (5N deviation), that is, It is within an acceptable range, but further research and development is still needed; in terms of taste and smell, it is revealed that Samsung’s “Semicon humanoid robot” plan will allow robots to have human-like smell and taste capabilities; as for hearing, current technology has basically achieved it its needs. In addition, the sixth sense in addition to the five basic senses requires continuous upgrading and iteration of the "brain and body" in order to have a chance to be realized.
The underlying technical logic of humanoid robots is constantly being overcome. Stimulated by the vast blue ocean market, companies have also begun to rush to lay out the humanoid robot track. At the 2023 Beijing World Robot Exhibition, participating in the exhibition of humanoid robot products There are more than 10 companies, and the summary of humanoid machine product parameters released by three companies is shown in Figure 1. It can be found that the "brain" of the humanoid robot at the software level and the "body" at the hardware level have achieved breakthroughs from 1 to N. Among them, more and more companies are beginning to shift their focus from motion planning to motion decision-making, that is, they are more inclined to develop the "brain" of humanoid robots. For example, Zhiyuan Robot's EI-Brain architecture combines the robot's The thinking system is divided into cloud super brain, terminal brain, cerebellum and brainstem, which implement different levels of instructions respectively, and continuously strengthen self-learning capabilities during the execution of instructions.
Since the creation of robots, people have never stopped imagining and developing humanoid robots. In the early stages of development, humanoid robots could only achieve walking functions and lacked more flexibility and intelligence. It could be said to be a body without a "soul". However, driven by the rapid development of technology, it is necessary to solve the shortcomings of humanoid robots. There are endless means, and the original impossibility gradually becomes possible. Nowadays, the basic development elements of humanoid robots have been gathered together, and the "robot revolution" that was once out of reach may be quietly coming!
This article is for knowledge sharing only and does not constitute the provision or reliance on investment, accounting, legal or tax advice. Any investment decisions based on this information are at your own risk.
The above is the detailed content of Robots move towards 'human'. For more information, please follow other related articles on the PHP Chinese website!

通常,机器人的主要功能是完成一些简单的操作任务,我们希望机器人可以模仿人,让能力尽可能接近人类水平。不论是小米的 CyberOne 还是特斯拉的 Optimus,人们关心的主要是其机械关节数量,控制算法和行走速度。不过在这个领域,有些人探索的方向更加脑洞大开:现在,有一种机器人把模仿真人表情做到了极致:先尝试一下自拍。从「嫌弃」到「惊讶」,都可以做到完全同步:这个机器人名叫 Ameca,是个表情怪。除了模仿,它自己也能照镜子做很多小表情,看起来非常像真人。Ameca「假装」第一次见到镜子,首

和活生生的已故历史名人聊天是个什么感觉?近日,就有一群开发者利用语言模型,把千百年来各行各业的历史名人全部「复活」成了聊天机器人,做进了一款手机app里,起名叫「你好,历史」!开发者声称,这个与古代名人聊天的app涉及的内容几乎无所不包。比如可以:与玛丽莲·梦露聊好莱坞八卦与弗里达·卡洛讨论现代艺术问问圣诞老人他有多少只驯鹿问问科特·科本为什么自杀向穴居人学习如何生火与宇宙意识辩论生命的意义不过他们也没忘记提醒用户,这些对话是由人工智能生成的,所以不要太认真。而且每个对话都是独一无二的,你永远不

有抱负的工程师应该了解世界各地著名的机器人工程学院。现在是从事机器人和工程事业的最佳时机——从人工智能到太空探索,这一领域充满了令人兴奋的创新和进步。美国劳工统计局估计,未来10年,机械工程领域的职业总体上将保持7%的稳定增长率,确保毕业生将有大量的就业机会。机器人工程专业的学生平均工资超过9万美元,无需担心还助学贷款的问题。对于那些考虑投身机器人工程领域的人来说,选择一所合适的大学是非常重要的。世界上许多顶尖的机器人工程学院都在美国,尽管国外也有一些很棒的项目。这是7所世界上最好的机器人工程学

大数据文摘出品作者:Caleb为庆祝英国女王伊丽莎白二世登基70周年,英国也是早早就洋溢出了庆典的味道。据了解,英国将于6月2日至5日连放4天公众假期,并在期间举行多项庆祝活动。英国皇家铸币厂也在精心打造有史以来最大的硬币,直径220毫米,重15公斤,面值15000英镑,耗时近400小时打造,是该厂1100年来生产的最大硬币。这枚金币一面雕刻着代表英国女王伊丽莎白二世的符号EⅡR,周围环绕着代表英国的玫瑰、水仙、蓟和三叶草。另一面有女王骑在马背上的图案。在这么热闹的日子里,AI当然也必须来凑一凑

人类与人工智能相比,哪个更擅长建立关系?事实上,这项革命性的技术已经存在了很长一段时间。然而,直到最近人们才意识到人工智能对人类的重要性。人工智能利用算法模拟人类,并随着时间的推移从经验中学习的能力,为这项技术与人类建立关系开辟道路。人类如何建立人际关系作为人类,我们倾向于只与少数人建立关系。我们试图确保不需要的和不相干的人从我们的生活中消失。在将我们的关系限制在少数人的同时,我们确保与那些对我们真正重要的人建立高质量的关系。然而,同样的方法在商业用语中可能不是理想的,并可能适得其反。尽管知道这

机器人也能干咖啡师的活了!比如让它把奶泡和咖啡搅拌均匀,效果是这样的:然后上点难度,做杯拿铁,再用搅拌棒做个图案,也是轻松拿下:这些是在已被ICLR 2023接收为Spotlight的一项研究基础上做到的,他们推出了提出流体操控新基准FluidLab以及多材料可微物理引擎FluidEngine。研究团队成员分别来自CMU、达特茅斯学院、哥伦比亚大学、MIT、MIT-IBM Watson AI Lab、马萨诸塞大学阿默斯特分校。在FluidLab的加持下,未来机器人处理更多复杂场景下的流体工作也都

还记得那个和特斯拉飙车的机器人吗?这是瑞士苏黎世联邦理工学院衍生公司研发的与公司同名的四足轮腿式机器人——Swiss-Mile,前身是ANYmal四足机器人。距离它和特斯拉飙车还不到半年的时间,它又实现了重大升级。这次升级改进了机器人的算法,运动能力直接UP UP UP ! 可以双腿站立下楼梯:(小编内心OS:如果是我穿轮滑鞋下楼梯可能会摔个狗吃屎)楼梯爬累了,坐个电梯吧,用前脚按开电梯门:面对障碍物应对自如:它还能知道什么时候该站起来,什么时候该“趴下”,双腿直立与四足运动之间的切换更丝滑:

随着人工智能成为一种趋势,人们对它如何工作以及可以做什么存在很多问题。一个经常被问到的问题就是——AI有哪四种类型。下面小编就来为大家解答一下。AI有哪四种类型反应性机器反应性机器在AI中是一个非常受欢迎的概念。这是因为它是最基本也是最古老的AI类型。反应性机器是指仅对某些刺激和场景有反应性的机器。与之后的许多人工智能软件不同的是,它们不能利用以前的经验或负载的知识来评估和应对特定的情况。甚至不使用GPS或数字地图来导航周围的环境或绘制路线。相反,它们会根据所看到的东西移动。反应型机器擅长象棋和


Hot AI Tools

Undresser.AI Undress
AI-powered app for creating realistic nude photos

AI Clothes Remover
Online AI tool for removing clothes from photos.

Undress AI Tool
Undress images for free

Clothoff.io
AI clothes remover

AI Hentai Generator
Generate AI Hentai for free.

Hot Article

Hot Tools

MinGW - Minimalist GNU for Windows
This project is in the process of being migrated to osdn.net/projects/mingw, you can continue to follow us there. MinGW: A native Windows port of the GNU Compiler Collection (GCC), freely distributable import libraries and header files for building native Windows applications; includes extensions to the MSVC runtime to support C99 functionality. All MinGW software can run on 64-bit Windows platforms.

SAP NetWeaver Server Adapter for Eclipse
Integrate Eclipse with SAP NetWeaver application server.

MantisBT
Mantis is an easy-to-deploy web-based defect tracking tool designed to aid in product defect tracking. It requires PHP, MySQL and a web server. Check out our demo and hosting services.

Dreamweaver CS6
Visual web development tools

SublimeText3 Mac version
God-level code editing software (SublimeText3)
