"Our study has tackled the intricate challenge of replicating efficient walking at various speeds - a cornerstone of the human walking mechanism," points out Associate Professor Dai Owaki, and co-author of the study along with Shunsuke Koseki and Professor Mitsuhiro Hayashibe. "These insights are pivotal in pushing the boundaries for understanding human locomotion, adaptation, and efficiency."

The achievement was thanks to an innovative algorithm. The algorithm evolved beyond the conventional least squares method and helped devise a neural circuit model optimized for energy efficiency over diverse walking speeds. Intensive analysis of these neural circuits, particularly those controlling the muscles in the leg swing phase, unveiled critical elements of energy-saving walking strategies. These revelations enhance our grasp of the complex neural network mechanisms that underpin human gait and its effectiveness.

Owaki stresses that the knowledge uncovered in the study will help lay the groundwork for future technological advancements. "The successful emulation of variable-speed walking in a musculoskeletal model, combined with sophisticated neural circuitry, marks a pivotal advancement in merging neuroscience, biomechanics, and robotics. It will revolutionize the design and development of high-performance bipedal robots, advanced prosthetic limbs, and state-of-the-art- powered exoskeletons."

Such developments could improve mobility solutions for individuals with disabilities and advance robotic technologies used in everyday life.

A schematic overview of the proposed optimization algorithm (PWLS). The plotted data points show the relationship between control parameter values and reproduced walking speeds. If the function expressing this relationship was obtained using the general least squares method, the data generating energy-efficient and non-energy-efficient gait are not distinguished and treated with the same evaluation. In the proposed PWLS, by adding weights based on evaluation values (energy efficiency) to the data points that generate energy-efficient walking, it is possible to obtain the relationship between the walking speed and the control parameters that realize highly efficient walking and to construct a neural circuit model that enables the generation of more energy-efficient walking.

Looking ahead, Owaki and his team hope to further refine the reflex control framework to recreate a broader range of human walking speeds and movements. They also plan to apply the insights and algorithms from the study to create more adaptive and energy-efficient prosthetics, powered suits, and bipedal robots. This includes integrating the identified neural circuits into these applications to enhance their functionality and naturalness of movement.