Turing robot CEO Yu Zhichen: 2045 will usher in the era of strong artificial intelligence

"In the same arena, every athlete is trying to run, but he does not know that some players have already gotten on the sports car. There is no doubt that this will run faster."

At yesterday's Turing Robotics Innovation Conference, Yu Zhichen, CEO of Turing Robots, commented on the company's development in the past year. According to the official conference data:

As of July 11, the third-party developer of the Turing robot platform exceeded 230,000, an increase of 110% compared with data released by Turing in October last year. At the same time, it has increased 16.2 billion times in the past 8 months. Request for service. In terms of technology, the accuracy of semantic understanding has been raised to 95.5%.

The reason why it can become the leader of domestic artificial intelligence, Yu Zhichen thinks that the “sports car” owned by Turing has three major advantages:

Powerful technology, Turing OS basically owns the core technology that our company has developed and accumulated over the years;

With stable services, artificial intelligence has many services that need to be processed in the cloud. Turing Robot has accumulated five years and has a 99.9% stability guarantee for the service.

The open system, through cooperation, can reduce the R&D costs of the entire product and shorten the product development cycle.

However, Yu Zhichen is not satisfied with this, in addition to the realization of the goal of “letting intelligent robots enter every family”, he also put forward a vision of strong artificial intelligence at the conference.

Strong artificial intelligence means to create intelligent robots that can truly reason and solve problems. Essentially, they are highly humanoid and can even have their own consciousness and sense of self-determination.

As for when this can be achieved, Yu Zhichen temporarily set the time to 2045.

The conference also released the Turing OS 1.5 version and dozens of robotics applications, covering education, entertainment, social networking, tools, and remote areas. Turing OS version 1.5 is enhanced in visual capabilities, motion control and hardware modules compared to Turing OS 1.0.

On the visual aspect, Turing OS1.5 will add 11 visual capabilities, including face recognition, face detection, face tracking and many other visual technologies.

In terms of motion control, Turing OS1.5 enhances support for 17- to 20-degree-of-freedom bipedal gait robots;

In terms of hardware modules, the Turing OS1.5 completes the motherboard and microphone arrays, and the laser radar is in beta.

According to reports, Doraemon robots equipped with Turing OS1.0 crowdfunded over 10 million, ranking first in domestic robot crowdfunding, and Le Di robots are also in full swing in preparation for sales forecast to exceed 100,000 units for 14 years. The AI+IP strategy developed was initially verified. At the same time, several robots equipped with Turing OS1.5 will be available in the second half of the year.

As for the characteristics and laws that robot applications should have in the end, Yu Zhichen summarized five points.

First, family. Family is one of the most important application scenarios of current intelligent robots. There are many interesting things that can be done around the family.

Second, autonomy. More mobile phones interact through the user's active trigger. The robot automatically performs tasks through the needs of specific scenes to meet the needs of users;

Third, multimodality. Multi-modality is similar to the mobile phone UI, and it must be considered when developing a robot system. Multimodality will combine limb movements, emotions, and other behaviors to embody a state when the robot speaks ;

Fourth, emotions. Emotion is also unique to robots. Robots can express emotions through expressions and body movements, and can also recognize emotions. Therefore, when designing robot functions, the consideration of emotions usually takes a lot of time;

Fifth, exercise. The robot has a very rich motion system, and this motion system must also be taken into account when designing robot movements. For this purpose, more interesting and optimized applications can be made.

At present, Turing OS is based on the Android operating system developed by the robot, but Yu Zhichen said in an interview after the meeting, Turing also has a set of Lunix-based programs, when the price-performance and Android-based systems are equal, Turing will be launched Native robotic system. How long will it take for concrete to be realized, and everyone will look forward to it together.

This entry was posted in on