It’s a micro robot driven by a magnetic field, and it’s hopping towards its target in motion. In a constant magnetic field, it can also give you a loop. Some people may think that the magnetic driven micro robot is not rare, even a little mediocre. However, the above robot is not ordinary, it has neurons and can form and manipulate neural networks between neural clusters in vitro. Therefore, it has achieved a new breakthrough in the research of brain function and related diseases. < / P > < p > this robot is jointly developed by the Korea Brain Institute and the Department of robotics engineering, dgist-eth micro robot research center and brain and cognitive science department of Daegu gyeongbei Institute of science and Technology (dgist). < / P > < p > for a long time, brain scientists have been trying to understand the brain’s learning, memory, movement, sensory processing and decision-making functions, and the realization of these functions in the brain is inseparable from neural connections. Scientists have proposed a method of neural network analysis to study the neural function in vitro. < / P > < p > the reason is that this method can make accurate and selective neural connections in the target position of the brain on the premise of reducing external influence as far as possible, so as to measure neural activity and determine the communication mode of neurons. Of course, it can also help understand axonal regrowth in injured or dysfunctional neurons. < / P > < p > in order to form and control the growth pattern of cell neurites, scientists all over the world have tried chemical, physical and mechanical methods, while scientists from Korea Brain Research Institute and Daegu Qingbei Institute of science and Technology (dgist) in Korea have designed a robot. < p > < p > at present, the existing research results show that the spherical, spiral and burr shaped porous spherical micro robots driven by magnetic field can achieve targeted cell delivery in vivo or in vitro. < / P > < p > previous studies focused on the fabrication of micro robots with various shapes and loading cells onto the micro robots under external power supply. To the best of our knowledge, no scientific team has reported the use of microrobots to regulate neurite alignment and neural connections. < p > < p > therefore, they designed a 3D magnetic driven micro robot with neurons, which can accurately transmit the neural network to the gap between two neural clusters through the external magnetic field, and then selectively connect the neural network. At the same time, the extracellular action potential is transmitted from one cluster to another through the neurons carried by the micro robot. < / P > < p > the above figure a shows the active construction of neural networks between two neural clusters. In this process, it mainly relies on a high-density multi-stage array chip embedded in the robot, which can measure the axon signal transmission. < p > < p > Part C shows the process of robot preparation using three-dimensional laser lithography technology based on two-photon polymerization (TPP) and deposition of nickel (Ni, for magnetic) and titanium dioxide (TiO2, for biocompatibility) layers. < / P > < p > control group (glass matrix): only a few cells were observed. The thickness of neurite is about 2-5 μ m, and the thickness of neuron body is about 10-20 μ M. < / P > < p > that is to say, compared with the control group, neurons can be successfully cultured with robots, and the neurite can grow without significant influence on the survival rate. < / P > < p > on the basis of neuron culture, this micro robot builds a neural network, and this process is realized by applying magnetic field to the robot on the neural cluster array. < / P > < p > the scientists designed a 20 MT and 1.2 Hz magnetic field through a linear superposition of eight solenoid hemispheres and a charge coupled device (CCD) camera on top of it. < / P > < p > in fact, in order to realize the active connection of neural networks, the key is to accurately transfer and locate the neurons cultured on the robot to the designated position. Although the cells attached to the robot add extra weight, which may affect the robot’s progress, scientists have used magnetic fields to achieve precise control with accuracy of tens of μ m (with an error range of about 10%). < / P > < p > as shown in the figure above, the neurons reach the target position in 10 seconds and align the clusters needed to connect the network within 1 minute. It is hoped that our research results will create a new breakthrough for the advanced artificial neural network controllable in vitro model. We are also using various micro robots to establish complex and diverse connections, hoping to improve people’s understanding of neural networks. It is said that “gta5” will be launched on Google cloud game platform stadia