A self-driving lab (SDL) was used to discover a structure with 75.2% energy-absorbing efficiency, surpassing previous records. The SDL performed over 25,000 experiments on generalized cylindrical shells (GCS) in an 11-dimensional parameter space, using Bayesian optimization to select experiments and human oversight to adjust parameters. The result was a structure with high energy absorption efficiency, demonstrating the potential of human-SDL collaboration in material design. The study also revealed transferable principles for designing tough structures, including the importance of material properties and geometric parameters. The SDL was able to identify high-performing designs in different materials, such as PLA and TPU-2, and showed that these designs could be adapted to other materials. The research highlighted the importance of material properties, such as elastic modulus and plateau stress, in determining energy absorption efficiency. The study also explored the mechanical behavior of cylindrical shells under compression, identifying key factors that influence energy absorption. The results demonstrated the effectiveness of SDLs in accelerating the discovery of high-performance materials and structures, and provided insights into the design of efficient mechanical systems. The study also emphasized the importance of human-machine collaboration in overcoming design challenges and improving the performance of energy-absorbing structures. The findings have implications for both mechanics and design, offering new strategies for creating efficient and safe protective equipment. The research underscores the potential of SDLs in materials discovery and highlights the value of combining computational and experimental approaches to advance the field of mechanical engineering.A self-driving lab (SDL) was used to discover a structure with 75.2% energy-absorbing efficiency, surpassing previous records. The SDL performed over 25,000 experiments on generalized cylindrical shells (GCS) in an 11-dimensional parameter space, using Bayesian optimization to select experiments and human oversight to adjust parameters. The result was a structure with high energy absorption efficiency, demonstrating the potential of human-SDL collaboration in material design. The study also revealed transferable principles for designing tough structures, including the importance of material properties and geometric parameters. The SDL was able to identify high-performing designs in different materials, such as PLA and TPU-2, and showed that these designs could be adapted to other materials. The research highlighted the importance of material properties, such as elastic modulus and plateau stress, in determining energy absorption efficiency. The study also explored the mechanical behavior of cylindrical shells under compression, identifying key factors that influence energy absorption. The results demonstrated the effectiveness of SDLs in accelerating the discovery of high-performance materials and structures, and provided insights into the design of efficient mechanical systems. The study also emphasized the importance of human-machine collaboration in overcoming design challenges and improving the performance of energy-absorbing structures. The findings have implications for both mechanics and design, offering new strategies for creating efficient and safe protective equipment. The research underscores the potential of SDLs in materials discovery and highlights the value of combining computational and experimental approaches to advance the field of mechanical engineering.