The German SME intends to develop and commercialize novel applications, products and services for forecasting of globally relevant, spatio-temporal problems in energy, climate, oceanography, health care, water supply management, infrastructure, and agriculture. The solutions are based on original software for parallel, self-organizing modeling and high-performance computing from observed spatiotemporal data.
The company developed a number of unique and innovative original technologies for validation of machine learning models implemented in commercial software products including new approaches for application domain identification, cost-sensitive and ensemble modeling with per-sample prediction uncertainty, and self-organizing knowledge extraction from high-dimensional state space. With the core technology behind, the forecasting is more exact then other exisiting models.
The startup company has established research co-operations with recognized scientists and experts in adaptive learning, control systems, and knowledge mining from institutions in London, Prague, Chengdu, Milan and cooperates with UNESCO.
It is currently preparing and developing a number of research and commercial projects in particular in the sectors of energy and climate change. Pilot users and partners are sought in order to complete and commercialize the product development according to different application scenarios under research, joint venture or commercial agreement with technical assistance.
Advantages & innovations
The forecasting tool is based on a core technology which enables high-resolution forecasting by means of a self-learning AI modeling. With the core technology behind, it is to be seen as pioneer by serving to different branches and scenarios.
The high-resolution forecasting system can be used for:
- Short to medium term climate forecasting includes temperatures, ozone concentration, cloud cover, aerosols, rainfall, risk assessment for insurances
- Short to medium term energy forecasting
of electricity, heat load, renewable power generation (wind, solar, selecting sites and operational planning), rainfall
- Short to medium forecasting in agriculture
of temperatures (crop yields), water resources and demand, droughts, floods
- Short to medium forecasting in oceanography
water quality (sea surface chlorophyll concentration)
- Forecasting in health
of temperatures and air pollution, UV radiation (ozone), cloud cover
More application fields are to be explored in frame of the partnership.
The SME has been awarded with grants from different innovation bodies, i.a. granted by the H2020 SME Instrument. Project collaboration related to the energy sector for the Indian market should be especially followed up.
The company has a motivated team of experts complementing each other with experience in original self-organizing machine learning technologies and high-performance multi-core parallel computing, business, management and consulting within global corporations and international organisations, including the European Commission, the European Center for Nuclear Research (CERN) et al., and with both academic and industrial experience in the areas of renewable energies in Germany and India.
The ambition is to become a major European player in R&D driven forecasting and self-learning predictive modelling technologies, leveraging the power of earth observation data, for highly relevant industrial and societal problems.
Stage of development
Project already started
Partner sought
The company has a project in mind and looks for pilot users or business partners to collaborate with. Partners can come from industry of different branches to adapt and commercialize forecasting applications which are globally relevant to solve spatio-temporal problems in energy, climate, oceanography, health care, water supply management, and agriculture.
Users are especially addressed from energy and climate protection sectors, but also health and other branches where forecasting models play a role are asked to participate in joint project activities for pilot user activities.