Is it possible for robots to learn how to successfully navigate the twists and turns of a  labyrinth? Well, researchers at the Eindhoven University of Technology (TU/e) in the Netherlands and the Max Planck Institute for Polymer Research in Mainz, Germany, have made it possible and proved that there is no such thing as “impossible” when it comes to technology.

However, machine learning, like every successful thing in this world, has its disadvantages. One of them is consuming too much human brain mimicry.

As we know, there are neurons in our brain that communicate with one another through so-called synapses. They are strengthened each time information flows through them. It is this plasticity that ensures that humans remember and learn, and researchers find inspiration in it in order to create a more efficient machine.  Imke Krauhausen, Ph.D. student at the Department of Mechanical Engineering at TU/e, explains:

“In our research, we have taken this model to develop a robot that is able to learn to move through a labyrinth. Just as a synapse in a mouse brain is strengthened each time it takes the correct turn in a psychologist’s maze, our device is ‘tuned’ by applying a certain amount of electricity. By tuning the resistance in the device, you change the voltage that controls the motors. They, in turn, determine whether the robot turns right or left.”

Krauhausen and her team created a robot with the help of a robotics kit, made by Lego. It is a Mindstorms EV3 and it is equipped with two wheels, traditional guiding software which is supposed to make sure it can follow a line, and a number of reflectance and touch sensors.

When put in a maze, the robot is told to turn to either return or to turn left every time it reaches a dead-end or diverges from the designated path to the exit. Krauhausen says:

“In the end, it took our robot 16 runs to find the exit successfully. And, what’s more, once it has learned to navigate this specific route (target path 1), it can navigate any other path that it is given in one go (target path 2). So, the knowledge it has acquired is generalizable.”

Organic material is used for the neuromorphic robot. It is both stable and able to maintain a large part of the specific states in which it has been tuned during the various runs through the labyrinth. This ensures that the learned behavior ‘sticks’, just like neurons and synapses in a human brain remember events or actions.

During the research, which dated from 2015 and 2017, the material was proved to be able to tune in a much larger range of conduction than inorganic and materials, and that it is able to ‘remember’ or store learned states for extended periods. Since then, organic devices have become a hot topic in the field of hardware-based artificial neural networks. Krauhausen added:

“Because of their organic nature, these smart devices can in principle be integrated with actual nerve cells. Say you lost your arm during an injury. Then you could potentially use these devices to link your body to a bionic hand,” says Krauhausen.

She also said that their robots rely on traditional software to move around. She admitted that she will be working on developing in the next phase of her research.