CELUS GmbH, a deep tech software company based in Munich, Germany says “yes!” and has developed an engineering platform to do just that. How? By developing a cloud-based engineering design platform that utilizes AI to create electronics designs quickly and intelligently as Tobias Pohl CELUS CEO and Founder describe as, “with the mind of an engineer” using their AI-driven engineering solution named Supernova.
What does AI for PCB design mean?
For companies that design electronic products, the path from product inception to the physical PCB design can be lengthy and cumbersome. Creating the design library elements including symbols and land patterns, making all the schematic diagram connections, and then placing and routing the PCB is both tedious and time-consuming. Now let’s imagine starting a new PCB design by simply creating a high-level block diagram, entering the functional specified circuit performance data required, and then letting an AI-driven design platform not only identify the qualifying components and circuits, but also create the bill of materials (BOM), the schematic diagram, and a PCB layout for you. CELUS’ AI-driven design creation platform was designed to do just that. That is, get designs from concept to 95% completion quickly such that design teams can take it across the finish line within their own native EDA Tools.
AI PCB Design Versus the Traditional Approach
Having spent many years designing PCBs myself it became evident early on that although PCB designs vary in density and complexity, etc. – that the flow, practices, and many of the steps taken during the design and layout stages are specific and standardized. That is, many physical characteristics of PCB layouts are often typical from design to design. For example, decoupling capacitors attached to power and ground pins of ICs and the location and circuit configurations of physical interface components like connectors and LEDs are typically near the board edge and so on. Many PCB designs also leverage design reuse with “golden” circuits and/or reference designs. So, given all the design aspects that are “known” about how schematics are connected and how PCB components are placed – it is completely conceivable that AI can be leveraged in many ways to create much of an electronics design.
What is CELUS’ Model for AI-driven Design?
The CELUS Engineering Platform was also designed to be fully integrated into existing electronics design companies’ EDA tool engineering environments. To do so CELUS developed an innovative AI-driven electronics product design tool named Supernova. Supernova was designed specifically to replace and accelerate the most often time-consuming and predictable manual steps during electronics design utilizing intelligent algorithms.
CELUS’ tools enable users to design products quickly and intuitively by selecting existing or user-defined high-level design circuit blocks named “Cubos”. Cubos are holistic circuit blocks containing all the necessary information, components, and specification data. Cubos are represented as single or hierarchical blocks that can be attached quickly with a single connection. Then, utilizing AI, the tool automatically processes the design information and creates the bill of materials (BOM), schematic, and PCB floorplan within minutes. Design data is fully compatible with leading EDA Tools. CELUS’ AI-driven design creation platform generally gets projects to 95% completion quickly to enable design teams to take it across the finish line.
How are Components Selected Using AI?
The CELUS Engineering Platform also removes the need for reading data sheets. Utilizing AI, it uses abstract engineering data to optimize and find the appropriate circuits and components for the design. Once identified the software will generate the BOM automatically – essentially turning user requirements into the complete component selections, schematics, as well as the PCB layout recommendation. In addition, engineers can also search and select preferred components and manufacturers based on their specific requirements. CELUS also enables users to import their own IP and know-how in the form of reusable modules, projects, components, templates, design rules, and specifications via their data management system named Orbit. Orbit is CELUS’ AI-sourced data management tool that provides a single source of truth for both component and design data library content through configurable roles, approval processes, and workflows.
CELUS’ AI Approach Advantages.
Utilizing AI, CELUS’ Engineering Platform combining the Supernova solution along with Orbit for library management was developed to extend the capacity and efficiency of electronics product development. Utilizing this AI approach enables teams to significantly shorten the design cycles and in turn, reduce costs and product time to market. Supernova’s UI cockpit was developed so that users can quickly design an electronic product by either creating new or placing and connecting existing functional circuit library modules onto the design canvas. Once the high-level diagram is defined the software will intelligently analyze the combined requirements for the entire design and search for a circuit module or combination of circuit modules that accurately fulfills the design’s functional requirements.
CELUS is pioneering a path to AI-driven and automated electronics design. Their solution automates the historically time-consuming tasks encountered while creating electronics products. With ready-to-use circuit modules, the platform uses intelligence to generate design schematics and well as proposing PCB floorplans. The solution supports design variations, and design reuse, and provides early feasibility insights and data generation for prototyping. I look forward to following CELUS as their products continue to evolve and see how electronics design companies embrace and leverage AI for hardware design.