Robots that develop the unique sixth sense, thanks to advanced research

by Nasreen Parvez
Apr 8, 2022


Continuous advancements in AI technology and research robots are now getting the unique sixth sense

What is sixth sense technology?

Some experts believe that people are born with a sixth sense. It is the sense of proprioception, which is the perception or awareness of the position and movement of one’s own body. This feeling helps coordinate our movements.

Solid-state sensors, traditionally used in robotics, are unable to register the high-dimensional deformations of soft systems, making this sophisticated sense difficult to replicate in robots. Embedded soft resistive sensors, on the other hand, have the potential to solve this problem. With the rapid advances in AI technologies and investigations, the discovery of new methodologies with a variety of sensory materials and machine learning algorithms, scientists are getting closer and closer to overcoming the difficulty of using this approach.

The generic coding for the integration of sixth sense technologies on the robot is done by multiple software. Sixth Sense technology is a representation of the concept of augmented reality. Sixth Sense recognizes the items in our environment and presents information about them in a real-time context. The user can interact with the content through hand movements thanks to sixth sense technology. Compared to text and image based user interfaces, this is a much more efficient method.

After the robot is built and the sensors installed, the next step is to integrate digital information into the real world by programming the robot to take image recognition input, transform it into a sixth sense robot and Python was used in conjunction with code from the Arduino IDE to complete this task.

How does a Sixth Sense Robot work?

In terms of smell and taste, robots with chemical sensors could be much more accurate than humans, but building in proprioception, the robot’s awareness of itself and its body, is much more challenging and is a major reason why humanoid robots are so difficult to get right.

Small adjustments can make a big difference in human-robot interaction, wearable robotics and sensitive applications such as surgery.

In the case of hard robotics, this is usually solved by placing a number of tension and pressure sensors in each joint, allowing the robot to find out where its limbs are. That’s fine for rigid robots with a limited number of joints, but not for softer, more flexible robots.

Robotics are torn between having a large, complicated array of sensors for every degree of freedom in a robot’s mobility and having limited proprioception skills. This challenge is being addressed with new solutions, often requiring new arrays of sensory material and machine learning algorithms to fill the gaps.

They discuss the use of soft sensors spread randomly by a robotic finger in a recent study in Science Robotics. Rather than relying on data from a limited number of places, this placement is similar to the continuous adaptation of sensors in humans and animals.

The sensors allow the soft robot to respond to touch and pressure in different locations, creating a map of itself as it twists into difficult poses. A motion capture system observes the finger as it travels around, and the machine learning algorithm interprets the signals from the randomly distributed sensors. After training the robot’s neural network, it can link sensor feedback to the motion recording system’s detected finger position, which can then be discarded. The robot looks at its own movements to find out what shapes its soft body can take and then translates those shapes into the language of these soft sensors.

The advantages of this approach include the robots’ ability to predict complex movements and forces experienced by the soft robot (which is impossible with traditional methods) and also the fact that it can be applied to a variety of actuators and sensors.

The application of machine learning allows roboticists to build a reliable model for this complicated, non-linear system of actuator motion, which is difficult to achieve by just calculating the softly expected motion of the bot. It also mirrors the human proprioception system, which relies on redundant sensors that fluctuate in position as we age.

Machine learning techniques are revolutionizing robotics in ways never seen before. Combining these with our knowledge of how humans and other animals perceive and interact with the world around us brings robotics closer to being truly flexible and adaptable and ultimately ubiquitous.

Share this article

Do that thing to share

Leave a Comment