Periodic Reporting for period 1 - snickerdoodle (An affordable edge-computing device for high-performance and power-efficient robotics automation)
Reporting period: 2019-06-01 to 2019-09-30
Collaborative robots for manufacturing automation are spreading across the industry, where they are used alongside humans to improve factory production efficiency and reduce hazards. In terms of units, it is estimated that the worldwide stock of operational industrial robots will increase from about 2,097,500 units at the end of 2017 to 3,788,000 units by 2021. Between 2018 and 2021, it is estimated that almost 2.1 million new industrial robots will be installed in factories around the world. Technological advancements in robotics, big data, machine learning, artificial intelligence (AI), and the Internet of Things (IoT) are propelling the evolution of manufacturing automation. The issue with the massive scale up of robotic automation is the amount of data generated as a result of the new robots, tooling, and functions used for manufacturing on a massive scale. This data if not processed and organized in a proper way will significantly delay the necessary volume which is projected during the continued development of the Industrial revolution 4.0. The snickerdoodle will allow for faster and more flexible programming of industrial automation robots. Snickerdoodle minimizes the time and effort needed for programming of automatic robots and widens their range of applications. Its field programmable gate array (FPGA) Intellectual Property (IP) cores are built to address a variety of markets and applications. Moreover, its full-stack module enables to accelerate computationally intensive algorithms by up to 100x compared to conventional microprocessors and GPUs, thus reducing processing rate and power requirements. By enabling edge-computing, snickerdoodle reduces data communication delays by 99% and decreases data transmission costs by 66%. It supports latency-sensitive and bandwidth-intensive data processing applications, enables up to 5x more efficient machine learning, and allows real-time machine-to-machine operations.
Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far
During the feasibility assessment krtkl evaluated the technical and economic feasibility of the project. The findings strongly support the viability of the innovation and were used as a basis to develop the activities for the Phase 2 project. The company identified HW and SW upgrades neede dfor intensive industrial application. As well, the company identified companies interested in participating in validation during the phase 2 project. A thorough understanding of the robotic market, in particular at European level and identification of the target countries: the global industrial robotic market was analysed together with the five countries representing the majority of sales; the European market has been broken down to its top markets and its main drivers have been identified. Finally, the definition of the business model, commercialisation plan and industrialisation strategy was defined; a two revenue stream business model has been reconfirmed; the commercialisation plan builds on the partnership with global distributors to reach Germany, Italy, and France, the biggest European robotic markets.
Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)
The expected outcome of the project is to successfully execute the market validation and demonstrate that the value of highly efficient edge computing for process improvement and optimization in automation.The full-stack system eliminates the need for complex, multi-component architectures, reducing time and effort needed for implementation. Its General Purpose Input-Output (GPIO) pins that can be connected to peripherals allow for easy implementation for different applications. Snickerdoodle's companion app allows for remote control, minimising deployment times of operators and costs.