Issue 108

The great benefits of AIoT

Augusta Bianca Ene
Digital Transformation Manager @ Centrul de Inginerie Bosch din Cluj


AIoT is an acronym that stands for the use of Artificial Intelligence (AI) within the Internet of Things (IoT).

The Internet of Things is not a new concept anymore. More and more intelligent products and services are being connected in order to create added value for their users. For example, 92% of all Bosch product classes are connectable. It is believed that there will be around 25 billion connected things by 2025 compared to only 14,2 billion in 2019.

So as the devices are being connected and start sending data, what do we do with all that data?

On the other hand, Artificial Intelligence algorithms need data to learn from; they are only as good as the data they've been trained on. So, the key factor in this context is the data, which is resulting from the use of intelligent, connected products or from the interaction between people and machines or between the machines themselves. By linking IoT with AI and machine learning, we can draw the right conclusions from these huge quantities of data and can react to these data during product engineering in seconds. We learn from the data and can thus improve our products and services on an ongoing basis.

We must admit that the heaviest investments in the sphere of AI are by far being made in the United States and China. While the major IT companies from US earn most of their money with data-based services, European companies have the opportunity to become global leaders with industrial AI applications. It is specifically in this field that they can play to their strengths — both the manufacturing of complex physical products and the evaluation of machine and product data. It is precisely this domain knowledge that Bosch combines with IoT and AI expertise.

Creating value by adding AI to IoT

The value creation of a product encompasses multiple stages in its lifecycle from the product development and manufacturing to the product usage by the customer. In the traditional value creation cycle, the feedback loop with the product manufacturer is very slow because it needs manual interaction. Data is collected only through triggered and targeted activities and in relatively small batches by means of market research or quality claims that might appear. This means that the product must be used for a certain amount of time until the feedback collection can be started. The feedback action then takes place for a defined period of time, the data is collected and at the end of the activity the batch of data is sent to be analyzed. The result of the analysis is limited by the data that is available. For example, the conclusions that can be drawn on developing new features are constrained by the questions asked during the market research.

By adding AI to the IoT, we create a closed loop value creation cycle that can even be automated, so it is much faster and thus enables us to be more agile and focus even more on the users. The driving force of the cycle is the creation of business value and it is sustained by a constant data flow, its life-force.

The life cycle of a product and the data creation and gathering starts already in the ideation phase, goes through all stages of its engineering - both hardware and software -, its manufacturing and logistics and its interaction with the customer.

The data needs to be collected, stored and processed in a structured manner. This is not an easy step as the data coming from so many different sources is very diverse in type, quality and frequency. We truly speak here of Big Data.

It is also vital that we ensure that users can keep control and maintain sovereignty over their data at all times and that these data are always protected.

The data is then processed using data science and machine learning algorithms and based on their findings we can close the loop by improving our products with this knowledge. In Bosch our core expertise lies in the physical products and technologies. Our big advantage is the domain knowledge that ranges from the automotive world to consumer goods and industrial or building technologies. This vast domain knowledge allows us to understand the incoming data on a deeper level, identify possible new or missing data sources and create new cross-selling business opportunities.

Let's take a look at a practical example from Bosch

Automated Valet Parking (AVP) is the world's first infrastructure-supported solution that provides an automatic drive-up and park service.

AVP allows the vehicle to be left at the drop-off area of the parking garage without further ado, and all the driver needs to do is to activate the smartphone app. This establishes digital contact with the parking garage, and the route to a free parking spot is computed.

This is when the Bosch technology in the parking garage takes charge. The pilot parking garage in Stuttgart has some 180 Bosch ceiling-mounted stereo cameras with integrated algorithms to detect objects and measure distances. They send their images and metadata to a server in a separate room in the building. Stored on the server there is a digital map, a simplified blueprint of the building broken down into tiny grid squares no larger than a few centimeters. The server aligns data from the cameras with individual grids, detects the positions of people, objects, and the car using parameters such as size and direction of movement, and identifies available parking spaces.

The server uses a predefined path stored on the digital map to guide the car through the parking structure which consists of many individual segments, each of which contains specifications such as driving speed and curve radius. The server calculates a driving command for each segment and sends it out to guide the vehicle segment by segment into the parking space.

The cameras monitor the driving corridor and its surroundings and detect unexpected obstacles or persons in the car's path so that the vehicle can react immediately. The car first slows down when an obstacle is several meters away, and then stops when the gap closes to four meters. It brakes instantly if a pedestrian, rolling suitcase or other obstacles come any closer than that. The driving commands are calculated several times per second in parallel on the main server and on a second server, and the command is actually sent to the car only if both computers give the go-ahead. Failing that, the vehicle stops immediately. This autonomous driving system is faster than the average driver and has multiple safeguards. Various monitoring programs constantly check all components and functions and are in turn supervised by safety software.

Bosch expects the authorities to approve the automated valet parking system in the P6 Parking Garage at the Stuttgart airport for mass production by the fall of 2021 and it is already talking to automotive manufacturers and parking garage operators in other countries too. With good reason, as one thing is universally true: nobody likes parking.

Future perspectives

AIoT shows great promise also for residential and mobility applications - for the "life" domain - but it has to be Aware of its surroundings, Autonomous in collecting, processing and transmitting data and Actionable meaning that the AI derived conclusions are turned into actions. Still, a singular solution offers minimal value for companies. A vibrant IoT ecosystem of connected solutions based on a user centric approach unlocks valuable potential for companies and consumers alike.

This is part of our day to day business at Bosch Engineering Center Cluj, where we work on all levels involved in such applications, from the development of smart sensors, electronic control and communication units, data pipelines, cloud platforms and data science and AI solutions for various use cases from the mobility sector and not only.




  • Accenture
  • BT Code Crafters
  • Accesa
  • Bosch
  • Betfair
  • MHP
  • Connatix
  • BoatyardX
  • AboutYou
  • Telenav
  • .msg systems
  • Grab
  • Colors in projects