The search for life beyond earth is extending to the outer planets of the solar system. GAMMA is a concept space exploration lander designed using experimental AI-based software to significantly reduce mass for future missions to the outer solar system.
The moons of Jupiter and Saturn show promising signs they may contain the ingredients for life. One way to know for sure is to put a lander on the surface of these remote regions of space. That presents new design and engineering challenges. Landers perform complicated functions in temperatures far below zero and withstand radiation levels thousands of times greater than on Earth. Perhaps most importantly, they must travel hundreds of millions of miles through space before they can start their on-site work. For the distant journey to the outer planets, reducing mass is critical. Every kilogram that can be cut from the lander payload translates to mass reduction to the launch vehicle itself—and an exponential reduction in the fuel needed.
GAMMA was designed using generative design technology, a new AI-based approach that generates design solutions according to goals and constraints set by designers. It enables design teams to quickly explore a broad design solution space while still being bound by manufacturing and performance requirements dictated by the team or environment. Resulting designs meet set objectives and can be constrained and optimized for a range of fabrication processes, including casting, subtractive, and additive manufacturing.
GAMMA uses generative design and advanced manufacturing to produce a unique lander design unlike any other. The structural components of the lander are co-optimized as an assembly to enable design modularity and different manufacturing processes to be used for different feature scales. All structural components use lightweight aluminum alloys from either a metal casting, subtractive, or additive process. When compared to a conventional design, a mass reduction of 35% was achieved for the main structural component of the lander. This component has a unique optimized hollow design only made possible using an advanced manufacturing process. 3D printed sand molds are combined with traditional metal sand casting to allow the hollow channels of the design to form a large scale monolithic structure.
GAMMA addresses numerous requirements including serviceability during assembly, vibration of the structure during launch, radiation exposure, and thermal changes throughout the journey. GAMMA is the most complex generative design to date, meeting some of the most rigorous engineering standards to answer one of the grandest challenges – the discovery of life beyond earth.
GAMMA is a concept space exploration lander designed using experimental AI-based software to significantly reduce mass for future missions to the outer solar system. Below we explain the motivation, process, solution and results of the project.
JPL tackles some of the most challenging design and engineering problems in the world. Using these problems as inspiration, Autodesk is collaborating with JPL to research and prototype new capabilities to advance the state of the art for generative design and enabling the rapid creation of design solutions.
Generative design has the potential to help JPL iterate on designs faster and generate designs that significantly reduce mass. The importance of weight reduction for future space missions is significant due to the sheer distance to be travelled. Every kilogram of mass that can be cut from the structural payload enables a critical increase in the scientific payload of sensors and instruments to search for life beyond earth.
Design and manufacture a lander from the ground up to reduce mass, while meeting the strict structural performance requirements set out by JPL. Utilize conventional, as well as advanced, manufacturing processes to fabricate the physical lander.
A generative design process was used which focused on establishing and iterating on a 'problem definition' that describes the requirements for the design. The problem definition for the lander is extremely complex and takes into consideration stress, compliance, frequency, assembly, harnessing, and radiation requirements. As the fidelity of the problem definition grew, generative design was a critical tool for quickly iterating to produce sophisticated design results.
A combination of research and commercial grade software was used, including Autodesk Fusion 360 and Netfabb. Manufacturing constraints are used to constrain the generative geometry to ensure compatibility with casting, subtractive, and additive manufacturing. The output of solid models also enables manual edits to be made for late breaking design changes. Over 300 design iterations and 100 days of computation time we used to produce the final design.
DESIGN RESULTS & MANUFACTURE
The external structure holds the main payload of the lander and connects through to the legs. The external structure is generated as a variable wall thickness hollow structure. Material is algorithmically distributed to ensure structural performance criteria are met while minimizing mass. The external structure achieved a mass savings of 35% compared to the baseline design. The external structure is optimized for the sand casting process and uses additive molds rather than conventional tooling. Sand casting with additive molds enables large, complex geometry to be manufactured as a monolithic casting; avoiding the need for heavy bolted interfaces.
The internal structure holds the payload boxes in a specific configuration that allows for ease of assembly and compact packing. The internal structure is a multi-component assembly that is co-optimized using generative design. The internal structure achieved a 'payload mass to structure mass ratio' of approximately 10:1. Meaning the structure is able to support approximately ten times its weight. This result exceeds the 5:1 ratio commonly found in conventional high performance components. The internal structure uses a combination of traditional subtractive machining and additive manufacturing.
The lander legs utilized subtractive manufacturing constraints by defining four machine setup orientations in the generative design environment. These constraints ensured the final design could be fabricated from stock Aluminum on a standard 3 axis CNC machine.
CONCLUSION As scientists and researchers look further out into the universe for signs of life in the outer planets, the ability to develop spacecraft that can travel extreme distances (350+ Million miles) presents significant design and engineering challenges. Experimentation with new technologies like generative design is necessary for continued progress in these areas. GAMMA shows, for the first time, how generative design can be applied to a complex design problem of this nature.
Just as the computational power of mainframe computers helped the space program reach new heights in the 1960s, technologies like generative design are creating new possibilities in space exploration today, enabling us to go further and learn more about our place in the universe.