Simulation Gives Unmanned Aerial Systems Development Its Wings

We learn from William Giannetti, Swift Engineering senior research and development engineer, how simulation can be used to develop UAVs.

Engineers need more and better computer-aided engineering (CAE) tools to design, simulate and analyze increasingly complex components and systems. Purchasing expensive licenses for infrequently used tools can be a tough sell, but the analysis tools can so significantly improve efficiency, quality and time to market that finding a solution is critical.

A pay-as-you-go licensing model offers significant advantages over more traditional ownership licensing. Users of the Altair open architecture CAE simulation platform can buy HyperWorks Units, tokens used to buy access to static, dynamic, thermal structural and explicit analysis tools within the HyperWorks environment. Users can also access specialized analysis tools through the Altair Partner Alliance to analyze electromagnetic systems, multi-scale composites, and perform computational fluid dynamics (CFD).  This token-based pricing model enables users to use the variety of tools they need rather than being limited to a smaller set of licensed tools they own.

Users can import geometry and models from multiple CAD and CAE packages and formats into the HyperWorks environment to integrate component designs from multiple vendors, suppliers and sources into a complete design, eliminating the need to redraw or redesign due to format differences. Consultants and others working with multiple customers and systems can quickly import customer designs for analysis on a common platform.

Development of unmanned aerial vehicles (UAV) is one area in which the broad capabilities of the Altair system provides significant benefits. The multiple materials and technologies necessary to design an aerial vehicle drives users to perform more complex and varied analysis.

UAV developed and analyzed with Altair tools. (Image courtesy of Swift Engineering.)

UAV developed and analyzed with Altair tools. (Image courtesy of Swift Engineering.)

William Giannetti, Swift Engineering senior research and development engineer, explored the capabilities of aforementioned software tools in a presentation at the 2017 Americas ATC West conference. Swift developed a new vertical launch UAV platform using the tools and conducted design studies to evaluate the effect of accidental impact from a tool or foreign object, ditching and crashes to quantify the robustness of the design.

A steel rod simulates the impact of a tool striking the wing. (Image courtesy of Swift Engineering.)

A steel rod simulates the impact of a tool striking the wing. (Image courtesy of Swift Engineering.)

Giannetti modeled a foreign object as a steel rod and impacted the UAV in various areas to evaluate the effect. After creating finite element analysis (FEA) explicit models to simulate the aircraft and a steel rod penetrator, he simulated and analyzed the effect of forcing the steel rod onto or into the aircraft in critical locations to determine key parameters such as the location of maximum compression strain for a given kinetic energy. This analysis provided insight into vulnerable components at higher risk for damage. For example, the tool impact analysis showed the ribs and spars are most prone to damage and quantified the maximum limit on tool mass to mitigate impact and reduce risk of structural damage.

Analysis of a tool impacting the wing. (Image courtesy of Swift Engineering.)

Once the FEA explicit model is built, various impact studies can easily be performed using the model in RADIOSS, the explicit nonlinear structural analysis solver in the HyperWorks suite. Giannetti’s analysis included a landing drop study, bird strike, wall strike, ditch and steel penetrator strike. The HyperWorks suite provides an ideal environment to try different methods to analyze and improve products.

Impact studies on the UAV (Images courtesy of Swift Engineering.)

  1. Forward leg drop test landing

  1. Bird impact
  1. Ditch

  1. Wall impact
  1. Steel penetrator

This flexibility and range of analysis requires significant computation, and users performing complex analysis with large point counts are likely to encounter computational limitations. In spite of significant increases in computational capacity, real-time analysis remains a goal for the future. Improved algorithms and increasing computer performance mitigate this constraint to a degree, but significant computational capacity is still required to perform this depth of analysis. For example, completing the UAV ditch analysis required more than three weeks on an eight-core workstation.

Altair offers local cloud servers for HPC. (Image courtesy of Altair.)

Altair offers local cloud servers for HPC. (Image courtesy of Altair.)

Altair offers a cloud computing option to mitigate computational constraints. HyperWorks Unlimited Physical Appliance is Altair’s local cloud-computing solution. This solution combines a partner’s HPC expertise with Altair’s simulation technology to provide an integrated solution. This solution combines HPC cluster technology, CAE software and workload management capabilities into a turnkey, local, on-site cloud-computing solution. The HyperWorks Unlimited Physical Appliance system is fully optimized to execute massively parallel applications such as CFD and FEA Solvers. Users have more control over their proprietary data with local cloud computing and can keep data within their own facilities and resources.

Users requiring higher computational capability can use Internet-based cloud-computing resources with the HyperWorks Unlimited Virtual Appliance. This virtual cloud-computing system can scale as needed to complete analysis in reasonable time. A per-node, per-month pricing model maintains affordability while providing computational capacity as required.

Altair’s HyperWorks CAE environment offers significant analysis capability at an affordable cost. Visit the Altair website for more information about HyperWorks.

Altair has sponsored this post. They have no editorial input to this post. Unless otherwise stated, all opinions are mine. —Ralph Sprang