A sensing solution for smart cities

5 mins read

The AMANDA project looks to advance sensing capabilities to more smart applications, as Dr. Charis Kouzinopoulos explains.

Technology is vital to solve many of the challenges that society faces today. Whether it is fighting climate change or feeding a growing global population, technology offers some of the best chances of finding a solution.

None of the challenges that we face are simple, and they generally require much more advanced knowledge than we currently possess. Their scale is such that they will take expertise in many different scientific and technical disciplines, more than all but the largest company is willing to invest in either time or money.

To help tackle those challenges, and also to ensure that Europe is at the forefront of developing the scientific know-how of tomorrow, the EU introduced the Horizon 2020 funding programme to support research and bring businesses together with academia and research institutions to combine expertise.

The scale of Horizon 2020 is enormous. Since the programme begun in 2014, it has distributed almost €80 billion to around 31,000 projects and grants. As well as helping European research, the EU has also used Horizon 2020 to assist in the implementation of its other high-level policy initiatives, such as Europe 2020 and Innovation Union. In the same way, Horizon 2020 has been important to the implementation of the European environmental research and innovation policy - the EU’s project to promote a more sustainable economy and society as a whole.

Implementing a Solution

A large part of that economic greening revolves around finding and implementing more efficient methods of operation. A good example of the potential of this type of efficiency is occupancy detection in smart buildings. If we can detect when a room or area is unoccupied, we can automatically switch the lighting, HVAC and other systems off to save energy and cut costs for the owners of the buildings. The same principles apply to the direct emissions from traffic.

All too often, however, there are unnecessary emissions where vehicles are at a standstill, or when they are driving around looking for a place to park. Intelligent traffic management could direct vehicles to the most suitable parking area, or keep traffic moving by suggesting alternative routes, reducing the overall emissions for each trip.

The computing power to implement these smart processes is already available through data centres. The key to their success is sensing. Engineers and software developers often use the term RIRO, which stands for rubbish in/rubbish out. If the processing element doesn’t get the correct information as an input, then it is impossible to get the correct output. These inputs also have to be delivered in time for corrective action to be taken. The system processor can be the world’s most complex AI chip in a remote data centre, or a small, low-power microcontroller at the edge of the network, but in either case the result is the same, accurate information is critical to the success of the system and the better the information, the better the outcome. 

The AMANDA Project

The AMANDA project, funded as part of Horizon 2020, looks to advance our sensing capabilities for smart applications.

The project develops an Autonomous Smart Sensing Card (ASSC) with the same dimensions as a credit card and a height of only 3mm. The ASSC’s compact size, self-sustainability and wide variety of sensors allows it to be used in a many different environments and applications, and its integrated intelligence provides the flexibility to adapt it to new requirements. It can easily be used for tagging people or assets, which would have been especially useful during the COVID pandemic to keep track of equipment, assist in social distancing, track patients and even monitor the environmental conditions in which medicines are stored. It also has all the features necessary to be quickly integrated into room sensing systems for automated control and safety.

The complexity of the AMANDA project means that it involves the complete research and development cycle, including developing and optimising new components. To incorporate the electronics into such a small space required highly miniaturised components, and since the ASSC could be left on location for long periods of time, it will need to use an extremely low amount of energy and be self-powered.

Long- and short-range wireless connectivity are also required to report readings back to a gateway or other ASSCs in the network.

To carry out the project, 8 partners that included research institutes, SMEs and larger companies were chosen from 6 European countries. The eight participants offer a combination of expertise that included manufacturing infrastructure, micro- and nanotechnologies, composites, architectures and software.

Activities in the project were broken down into eight work packages, to be implemented during its three-year duration. The work packages take the project from system and architecture specification through to a working prototype. Sensor technologies have been developed from the beginning where optimised off-the-shelf technologies were not available. For the operation of the ASSC, a large variety of sensors is required. These sensing devices include environmental sensors to measure CO¬, volatile organic compounds, temperature and humidity. A spintronic sensor, low-power accelerometer and imaging sensor are required for general applications. Finally, sensors that allowed human interaction with the ASSC are included, for example a capacitive sensor, microphone and imaging sensor to allow authentication, recognition and activation. The sensors will be able to act both individually and in cooperation with other sensors to provide a device that offered true sensor fusion.

To support the sensing capabilities of the ASSC, further new technologies have been researched, developed and integrated if they were not available commercially. These technologies include energy storage, energy harvesting, power management, security and communications.

The direct responsibilities of the partners for the on-board technologies are:

  • The Centre for Research and Technology-Hellas (CERTH, Greece) is responsible for the coordination of the project, the integration of the hardware technologies on the ASSC, as well as on several software modules, including edge intelligence and security
  • Imec (Belgium) contributes the CO2 sensor
  • The ZHAW Zurich University of Applied Sciences (Switzerland) is in charge of the RF module for Sigfox, LoRa, and LoRaWan communications.
  • Lightricity (UK) is responsible for the photovoltaic energy harvesting device
  • e-peas (Belgium) has responsibility for power management and image sensing
  • Ilika (UK) develops the energy storage device
  • Microdul (Switzerland) contributes to the project with the temperature sensor and the capacitive sensor
  • Penta (Croatia) is the end user: at the end of the project, Penta will provide a report on the evaluation results, lessons learned, and key improvements that will optimize the system or increase the targeted spectrum of applications.

As well as those direct responsibilities, the members are also working together on other competencies, such as manufacturing, systems integration, software development and testing.

Final deliverables

As the majority of the project took place during the COVID-19 epidemic, its timeframe has been extended by 6 months until June 2022.

Despite the interruption caused by the pandemic, all of the work packages have been carried out successfully and a size unconstrained prototype was delivered and validated in terms of functionality and power efficiency. All of the components and building blocks of the architecture conform to the system specifications, along with the firmware.

The 8 partners will continue to work together for the remainder of the project to deliver a final device that conforms to the specified form factor. Once the project is finalised a fully autonomous and miniaturised sensing platform will be developed which will contain state-of-the-art sensors, local data processing capability and wireless connectivity.

The ASSC will be able to fit in your pocket for wearable applications but can also be installed in building or in outdoor applications.

By installing custom developed firmware the platform can be easily adjusted to suit a wide variety of applications.

Penta will incorporate the ASSC into company’s SMART ECO PARKING system (detection of parking space availability in real-time, helping to optimise on-street parking in cities and parking garages or surface parking lots such as those in shopping malls, bus and train stations etc.) The sensing capabilities of the ASSC will provide the information for Penta to more easily monitor capacity and provide more advanced payment and security services.

Author details: Dr. Charis Kouzinopoulos, Postdoctoral Researcher, Information Technologies Institute - CERTH, Greece