MITSUBISHI ELECTRIC Changes for the Better

Compact AI will change the world in big ways.

When AI goes to work in every part of our world, it will change the world in amazing ways. Mitsubishi Electric has made AI more compact so it can go on almost any device and makeeverything smarter.

Introducing Maisart: Mitsubishi Electric's AI with endless possibilities

Maisart stands for 'Mitsubishi Electric's AI creates the state-of-the-ART in technology'. The company makes the most of AI technology to make devices smarter and users' lives safer, more intuitive, convenient and functional.


Artificial Intelligence is a technology that uses computers to perform intellectual functions like logical inference or learning from experience, just as humans do. AI has evolved rapidly in recent years as computing devices have reached higher levels of performance. Nowadays, AI is an important technology supporting our society. Machine learning is one field of AI, and deep learning is one type of machine learning.

Deep learning is based on neural networks, which reproduce the network of human brain neurons as a mathematical model. A neural network is composed of three kinds of layers; the input layer, the hidden layer, the output layer.By processing information in multiple layers, neural networks are capable of high-level recognition, identification, analysis, etc. There is a great expectation that this technology will make computers more like humans.

Less calculation for the same inference accuracy.

There are several issues that need to be addressed for deep learning to become more widespread.One such issue is the great amount of calculation. It can be a challenge to equip factory automation, automobiles, and other equipment with deep learning because it is so hard to put high-performance computing devices and high-capacity memories on small devices. Mitsubishi Electric has developed a proprietary algorithm that greatly reduces the amount of calculation while maintaining a high level of inference accuracy.

The input, hidden, and output layers of a neural network connect to each other in complex ways, like tree branches spreading out. A massive amount of calculation is required to process data this way. Drawing on our machinery knowledge built up over many years, we successfully compacted the amount of calculation to just 1/30 to 1/100 the original amount by “cutting" what is less essential.


Reinforcement learning is a type of AI machine learning. Computers usually act following a human-created program. With reinforcement learning, however, a computer can understand the current situation by itself, set its own rules, and determine what action to take. Humans do not need to set the rules with a program. For a computer to determine what action to take next, it needs a lot of experiences, including experiences of failure, just as humans do.When we teach a robot some action, tightening a screw, for example, we make it try that action again and again. This is how it learns.

During reinforcement learning, a computer makes repeated attempts at actions and is evaluated (rewarded) based on how well it achieved the objective. It revises its action to get a higher evaluation, gradually getting closer and closer to the objective. Reinforcement learning is the part of AI that learns through the principle of “practice makes perfect”. It is the part of AI that finds success from failure.

Reducing the number of pre-learning trials by estimating degree of success.

Reinforcement learning does not require a human to set rules with a program. However, learning can take a lot of time because a huge number of trials are needed for pre-learning. Mitsubishi Electric has developed proprietary technology that reduces the number of trials to about 1/50 the conventional total. Conventional reinforcement learning senses trial results and sets control parameters based on evaluation of the same. In addition to that, Mitsubishi Electric’s technology uses our knowledge of the machinery that incorporates the AI to estimate the degree of success of trial results and sends feedback to the AI on what motions would get the equipment close to the target state faster. Control parameters are then set accordingly. This allows learning with fewer trials, making it possible to greatly reduce the time and cost of implementing AI.


Big data is broadly divided into data generated by humans (on social networks, for example) and data generated by things (such as sensors placed on equipment). The amount of data generated by things in particular is increasing rapidly as IoT spreads. Much attention is focusing on edge computing as a way to process this data quickly. This is because it would be very difficult to keep up with the explosive growth in data by existing means that depend on the cloud for all data processing, because that increases the data communications load and lowers responsiveness.

Edge computing can minimize the communications load and speed up processing by placing servers around the device and dispersing the processing of the data. In addition to dispersed processing by edge computing, it is important to have high-level big data analysis technology like AI in order to use big data effectively.

Drawing on our machinery knowledge to efficiently analyze time series data.

There are many ways to use big data, one of which is the preventive maintenance of infrastructure and the like. Promptly finding signs of irregularities in equipment requires fast and accurate analysis of vast amounts of time series data from sensors. Mitsubishi Electric has successfully used machine learning to reduce the number of calculation repetitions needed to detect signs of irregularities to just 1/40 the earlier count.

The technology until now would find signs of irregularities by comparing all the waveform data from sensors. Our technology makes it possible to find signs of irregularities quickly, with fewer calculation repetitions, by categorizing and learning waveform data in a number of typical patterns (clusters) and extracting only the differences (degree of deviation) in those patterns. Through AI and other analysis technology using Mitsubishi Electric’s machinery knowledge, we are helping to make time series data analysis faster and more efficient in edge computing.

Discover our solutions!


Artificial intelligence enables predictive maintenance functions, based on the real absorption of the robot axes. The MELFA FR family, characterized by high performances in terms of speed and repeatability, is able to predict potential failures, before they happen. Intelligence, Integration and Safety also define the pillars of the product range and embody the vision of robotics according to Mitsubishi Electric.

Thanks to Maisart technology, the rapid force control algorithm developed for the new force sensor, reduces operation times and eliminates irregular movements of the robot. Artificial intelligence allows to quickly adjust the process parameters, ensuring a cycle time reduced by 65% for insertion and assembly operations.

Inverter FR-E800 and Fault Diagnosis

The new intelligent functions allow to identify in advance potential faults of inverter and external parts and identify the causes, minimizing the system down-time. The new family of FR-E800 inverters with on-board Ethernet and advanced Safety functions also allows data collection and real-time monitoring of consumption variables, even remotely.

Servo MR-J5 and Predictive Maintenance

Thanks to the new functions, the servos of the MR-J5 family monitor the operating status of the machine and are able to detect possible anomalies in advance. In addition, the new servos with a 3.5 kHz bandwidth and battery-less absolute encoders (26-bit resolution), with over 67 million pulses per revolution, ensure high performance and cycle times of only 31.25 μs.

e-F@CTORY Starter Package

It is a set of basic programs integrated into the iQ-R modular PLC to support IoT systems in manufacturing. Ideal for monitoring production, developing predictive maintenance systems and managing quality control. Thanks to advanced statistical methods, the system provides visualization and continuous production management functions on GOT, allowing the user to visualize dashboards and graphs with the process KPIs.

Have a look at our AI videos

AI Storytelling: Mitsubishi Electric narrates the Artificial Intelligence

How can AI revolutionize the world?

The advent of Artificial Intelligence has revolutionized the world as we know it, new paradigms have overwhelmingly entered our private and professional lives, profoundly changing the dynamics of the industrial world and beyond.

Click  here  to download the first episode of AI Storytelling.

Do you know how to benefit from data usage?

"By far, the greatest danger of Artificial Intelligence is that people conclude too soon that they understand it," says noted researcher and writer Eliezer Yudkowsky, so continues our journey of discovery of AI.

Click  here  to download the second episode of AI Storytelling.