Visual data feedback: broaden your company’s vision

Share

Visual data feedback in machine learning processes machine learning has enabled the creation of various automated systems such as CUE4, the basketball robot. But What are the challenges ahead for Data Visualization Feedback if it wants to make this type of entities more and more human and/or better replicate reality?

What is visual feedback?

Any system, object or platform that seeks to interact with the environment or with a user has a visual output, for example: an ERP, a robot or a video game.

In the most advanced cases, these entities are able to perceive information from the real world, contrast it with their databases and finally act accordingly, i.e. provide a physical and/or virtual response.

This back and forth of information is also known as visual data feedback. And their effectiveness is enhanced since the devices have built-in cameras and can record their environment in real time.

The goal of visual feedback is to make the connection between the brain/chip and the body/structure more efficient. Thus, a system with this technology can improve its motor skills, since it acquires greater awareness and control of its movements and the surrounding environment.

<< Collaborative artificial intelligence: sum of chips and neurons >>

In some cases, such as augmented reality video games, this technology takes aspects of the real world and digitizes them into images that are replicated ad infinitum. This is similar to the effect that can be created with two rows of mirrors facing each other with an object in the center.

Data in motion

Initially CUE4 was an extra project of Toyota engineers and technologists, who in their spare time built this 6-foot tall robot to be able to shoot baskets like a professional basketball player.

Soon after, the Indonesian-based automotive company realized the potential of this peculiar basketball player, so it decided to support the project and make him one of its stars.

To shoot the ball , CUE4 uses a set of sensors distributed on its torso, which measure the distance to the basket and evaluate its position on the court, and from this they adjust the speed, height, force and angle of the shot.

He also has sensors on his feet to recognize balls, so he can pick them up to make his baskets. How well does he do in this regard? Better than any shooter in the NBA. The robot’s shots demonstrate the accuracy of its sensors and its ability to adjust its decisions based on the data it receives at the time.

During several of its exhibitions this robot has had a 62.5% three-point shooting percentage.which exceeds the records of players such as Kyle Korverwho last season had a shooting percentage of 53.6% from that same distance, and Stephen Currythe best long-distance shooter in the world today, who has made 43.6% of his three-pointers in his career.

Although for now CUE4 cannot run, dribble or execute the other fundamentals of basketball, it is a tangible example of the potential for visual feedback in business.

[embedded content]
CUE4 in action

Visual feedback challenges

This technology has presented advances in several industries and fields of knowledge, in addition to entertainment. Such is the case in medicine, where it is used to facilitate the retraining of patients’ movement, as described in this article.

However, visual feedback-based systems have several challenges to overcome, the most important of which are:

  • Lack of organized and well-stocked databases
  • The cost of artificial intelligence
  • Companies’ lack of experience with machine learning
  • Simulate the functioning of the neural networks of the human brain.
  • Translating its virtual power into more flexible, sustainable and autonomous structures

In fact, according to forecasts by the CUE4 developers themselves, it will take at least 20 years for robots and non-human entities to incorporate skills that allow them to mimic humans in a fluid way, especially in the mechanical aspect.

In your company, what would be the use of a visual data feedback system? Have you automated some of your production processes? What do you see as the biggest benefit of this technology?

Comment in the space below and subscribe to my blog to learn more about machine learning for business, as well as other topics of innovation and scientific technology applied to business.

Originally published in Jorge Pérez Colin Blog

Leave a Reply

Your email address will not be published. Required fields are marked *

Share

Related posts