Intelligent new battlefield, intelligent mobile robot with social attributes

2019-08-06

With the rapid development of new technologies such as intelligent control, navigation and positioning, and multi-sensor information coupling, the intelligent trend of robot products has become more and more obvious. Intelligent mobile robots with functions of perception, recognition, decision-making and execution have become a new battlefield for industry competition.

1. Intelligent system with social attributes

This thing starts with an intelligent system with social attributes. With the increasing awareness and mobility of intelligent systems, people and intelligent systems are increasingly integrated, and there is a human-machine integration environment. In this environment of human-computer integration, an intelligent system with social attributes is naturally developed. Since this kind of intelligent system has social attributes, it means that it has the ability to evolve. The following two aspects explore the evolvability of this system.

One aspect is that intelligent systems must have the ability to evolve internally. If the system is to be internally evolved, it is impossible to achieve through a single computing platform, so it is necessary to realize the interaction between different platforms, such as the interaction between the computing platform of the robot and the cloud, that is, the "cloud robot" concept.

Another aspect is the ability of intelligent systems to evolve externally. For example, the ability of an intelligent system to have external evolving capabilities through interaction between humans and robots is what we call the concept of "social robots."

2.the cloud mobile robot for long-term autonomous positioning

Cloud mobile robot for long-term autonomous positioning. This key technology also has three major challenges:

●The first challenge is to cover a large area, which requires a lot of data to describe the different environments;

●The second challenge is to adapt mobile robots to dynamic scenes, such as detecting and tracking static or dynamic objects, and learning more to predict environmental changes;

●The third challenge is that as mobile robots need to run for a long time, as time goes on, it requires more and more data storage.

  In this case, we need to make the robot have more storage space and more computing power, if it is only relying on their stand-alone body is actually very difficult to achieve.


  This is the concept of "cloud robot", which is the use of robot-side computing and cloud computing for robotics research.

3.social mobile robots with the ability to learn pedestrian behavior

    We know that robots that are studied using traditional methods use people in the environment as obstacles, which creates a problem: the overall system is inefficient to move, and is likely to fail even in crowded environments.


   Therefore, a research team has proposed a new type of intelligent mobile robot that has the ability to learn pedestrian behavior patterns in an unknown environment, thus enhancing the intelligent understanding and intelligent decision-making ability of mobile robots. The team did three things: the first is the unknown environment, the domain is perceptible, the second is the semi-dense environment semantic map production, and the third is the simultaneous positioning and map update in the human-machine integration dynamic environment.


   This system has two main contributions: Firstly, a robot-based regional mapping algorithm based on pedestrian behavior learning is proposed, which is called "Learning from Social Behaviors (LSB)". Secondly, it is adapted to the human-machine integration dynamic scene. The robot simultaneously locates and maps the method of construction (DM-LSAM). The focus of both is on how to strike a balance between dynamic environmental adaptability, positioning accuracy and robustness.



Market space analysis of China's intelligent mobile robot industry

   The global service robot market will reach $5.25 billion in 2018 and will grow rapidly to $15.69 billion in 2020. In 2018, the global market for home service robots, medical service robots, and public service robots is expected to be $4.48 billion, $2.54 billion, and $2.23 billion, respectively, with the home service robot market accounting for up to 48%. , respectively, 20, 24 percentage points higher than home service robots and public service robots.


   In 2017, the number of financing in the robot industry exceeded 110 times, and the total financing scale exceeded 20 billion yuan. The number of financing in 2018 is expected to exceed 2017. From the perspective of subdivision, there are many investment and financing cases in the fields of personal/home robots, AGV, collaborative robots, machine vision, etc., and the scale is large.

    Intelligent mobile robots will enter the 4.0 era, distribute the cloud brain in various places from the cloud to the end, and make full use of edge computing to provide more cost-effective services. In addition to the cognitive ability to achieve intelligent collaboration, intelligent mobile robots also have understanding and decision-making. The ability to achieve autonomous services.


Source of information: Ofweek Artificial Intelligence Network