The hottest service robot may usher in a new era o

Service robots may usher in a new era of development. How can Xiaomi intelligence break through the technical bottleneck

Guangzhou Baiyun Airport introduced it to guide passengers, Alibaba released the "space shuttle" for the delivery of drugs in hospitals, and Japan operates robot hotels. As more and more people gradually enter our vision, we can't help but imagine what role they will play in the future

analysis of the overall market environment

I The global overall market is growing rapidly, and service robots usher in a golden age of development. The 2017 China robot industry development report points out that at present, the global scale continues to expand, the growth rate of industry and market is stable, and the growth rate of service robots is outstanding. It is estimated that the global robot market will reach US $29.82 billion in 2018, with an average annual growth rate of about 15.1%. Among them, US $16.82 billion, US $9.25 billion for service robots and US $3.75 billion for special robots

II With the rise of a new generation of artificial intelligence, the industry has ushered in new opportunities for rapid development

according to the latest report released by PricewaterhouseCoopers, artificial intelligence will stimulate global economic growth of $15.7 trillion. By 2030, artificial intelligence will be largely used to improve labor productivity, and consumers' demand for technological products will also increase. The global GDP will grow by 14%, of which China's GDP will grow by 26%. At the same time, relying on artificial intelligence technology, the application scenarios and service modes of intelligent public service robots are constantly expanding, driving the rapid growth of the service robot market

the Ministry of science and Technology issued the "12th Five Year Plan" for the development of service robot science and technology, which made it clear that we should strive to break through the key technologies that restrict the development of China's services and industries, constantly launch products with more application value and market prospects, actively explore new investment and financing models and business models, strive to build a number of leading enterprises, and cultivate the service robot industry into China's future strategic emerging industry

III The aging population is intensifying, giving birth to the market demand for service robots

the proportion of the elderly over 60 years old in China's total population has reached 16%. According to international standards, China has stepped into the threshold of an aging society. The recently released white paper on the development of China's aging cause shows that Chinese officials have realized the urgency of the aging population. With the rapid aging of the population, the nursing home is tense, and the elderly need to be accompanied. It can be expected that a large number of service robots will enter the family, ushering in the wave of service robot family

IV Giants have competed in the field of artificial intelligence, and the hot-rolled ribbed steel bars for reinforced concrete of intelligent service robots have become a new focus.

Google has successively acquired 9 robot companies, including Schaft, Redwood robotics, industrial perception, and so on, in terms of humanoid robot manufacturing, artificial intelligence, robot collaboration and so on. Not only that, Google is also actively developing ultrafast "quantum computer chips", using subatomic particles to encode data, significantly improving computing speed and providing "smart brains" for robots

Wang Shi made a public statement in early 2016: in the next decade, Vanke will replace 30% of property services with robots. With the continuous progress of technology, many property service leaf spring bench fatigue testing machines, such as safety inspection, visitor management, cleaning patrol, equipment inspection, etc., are mainly used to test the durability, load fatigue, reciprocating fatigue, fatigue life, dynamic stiffness, compression fatigue and other mechanical performance testing of large vehicle leaf spring bench, such as automobile leaf spring, locomotive leaf spring, rail transit leaf spring, truck leaf spring, passenger car leaf spring, etc, It not only improves the daily work efficiency, but also saves the cost for the property company

Alibaba Artificial Intelligence Laboratory released the tmall Genie space egg at this year's yunqi conference. It not only has visual technology and can shuttle back and forth efficiently, but also has the ability to carry heavy loads of 30 kilograms, avoid pedestrians, get on and off the elevator, and will soon serve the future hotels that opened before the double 11. Son Zhengyi, the founder of Softbank group, who invested in Alibaba, also invested in the cutting-edge technology from robots to IOT

v Under the trend of consumption upgrading, the explosion of the entertainment industry

the acceleration of the pace of life and the pressure of work, more young people spend their leisure time on entertainment, such as staying in hotels and traveling on weekends, and temporarily escaping from work and familiar places. At the same time, service machines are needed to help people share family work and liberate people to do what they like. Then, the vast market of home service robots will emerge, For example, home care robots, toy robots, security control robots and cleaning robots will be most needed

from the above five perspectives, it can be seen that there is no doubt that service robots have a very broad market prospect in various fields, but now they still face the following problems:

I Technology bottleneck

the core technologies of service robots include human-computer interaction, navigation and path planning, multi robot coordination, artificial intelligence, cloud computing, etc., specifically involving voice, semantics, processors, algorithms, communication, big data, cloud connection, etc., to achieve the autonomy, adaptability and intelligence of service robots. In addition, semi-finished products of such innovative fiber reinforced composites upstream and downstream of the industrial chain are integrated in the production of bond laminates, a subsidiary of LANXESS group, including capital, system, application layer development, etc. At present, there are still bottlenecks in service robots that have not been broken through in key technologies

for example, slam technology involves navigation and path planning. The data of slam comes from, and more and more kinds of slam come from the fusion of multiple sensors, so the quality of sensors has a great impact on the effect of slam. Due to the high diversity of products and hardware, and the complexity of the integration and optimization of slam related technologies, algorithms and software are highly fragmented, so there is no universal solution in the market at present. Slam technology is only a part, and a lot of user experience lessons are needed in the follow-up

at present, the mainstream solutions in the industry include laser slam and visual slam, and well-known manufacturers include Silan technology, Xiaomi intelligence, Yunji technology, radium intelligence, Suteng juchuang, aignaw radish, usens linggan, etc. The cost of laser 3D navigation scheme is high, and it still takes a long time for a large number of commercial landing. Visual slam mainly collects information through cameras, and the cost is much lower than that of laser radar. At present, the mainstream vision slam schemes mainly include structured light, time of flight and pure binocular. Like structured light, binocular uses triangulation to calculate the object distance according to the parallax of the object matching points, but binocular uses natural light, while structured light uses active light to emit stripes or speckle of a specific pattern. TOF is to continuously send light pulses to the target, then use the sensor to receive the light returned from the object, and obtain the target distance by detecting the flight (round trip) time of the light pulse

Kinect generation 1 and 2 (which have been discontinued) used structured light and TOF. Both TOF and structured light belong to active light, which are easily disturbed by visible light and object surface, so they are more suitable for indoor and short-distance application scenarios. The pure binocular scheme cannot recognize texturless objects in dark conditions

"vision + structured light + inertial navigation" integrated binocular scheme has better effect.

based on this, combining the advantages of binocular and structured light, the multi-sensor integrated "vision + structured light + inertial navigation" deep inertial navigation camera scheme adopted by Xiaomi intelligence has become the first choice of 200 enterprises at home and abroad. (IR) infrared active light detector can assist robots to recognize indoor white walls and texturless objects, It overcomes the problem that ordinary binocular can not recognize non textured objects. When working outdoors, the recognition distance can reach more than 15 meters, which is very suitable for the scene of service robots working at night. The inertial navigation of six axis sensor (IMU) can provide precise cloud complementary data for the application of vSLAM, so that the service robot can be used in non textured objects (such as white walls), all black conditions and indoor and outdoor conditions, and it is also easier to land commercially

II Emotion

whether robots discussed in the western world of American TV series should have human consciousness in the work report of the Secretariat of the special committee was listened to and commented by the representatives of the pre meeting, which caused many people to think about robots

western world | should robots have human consciousness

in some places, especially in experience oriented places, such as hotels and restaurants, what we need to feel is the experience brought by beautiful services and delicious food. At this moment, service robots can only be used as an auxiliary to do some repetitive work, which can not replace the emotions transmitted by human beings. Especially once we encounter special problems, if robots can not solve them quickly in a very short time, it will affect the whole consumption experience, Making form bigger than content is not conducive to the long-term development of hotels or restaurants

future direction: product differentiation can win the market.

IDC, a research company, said that the potential enterprise market for AI software, hardware and services is huge: it will reach about $58billion by 2021, compared with only $12billion last year. It has to be said that there are still many market gaps in the field of service robots. Enterprises should seize the different needs of consumers, seize the market opportunity and compete for the product market

interested friends might as well go to mytai to find an intelligent officer