Green sea turtles (Chelonia mydas) can swim up to 50 km per day while only consuming seagrass or microalgae. How the animal accomplishes this vast journey on such low energy intake points to the effectiveness of their swimming technique and is a testament to the power of evolution. Understanding the green sea turtle's ability to accomplish these journeys requires insight into their propulsive strategies. Conducting animal testing to uncover their propulsive strategies brings significant challenges: firstly, the ethical issues of conducting experiments on an endangered animal, and secondly, the animal may not even swim with its regular routine during the experiments. In this work, we develop a new soft-robotic sea turtle that reproduces the real animal's form and function to provide biomechanical insights without the need for invasive experimentation. We found that the green sea turtle may only produce propulsion for approximately 30% of the limb beat cycle, with the remaining 70% exploiting a power-preserving low-drag glide. Due to the animal's large mass and relatively low drag coefficient, losses in swim speed are minimal during the gliding stage. These findings may lead to the creation of a new generation of robotic systems for ocean exploration that use an optimised derivative of the sea turtle propulsive strategy.

Sea turtles gracefully navigate their marine environments by flapping their pectoral flippers in an elegant routine to produce the required hydrodynamic forces required for locomotion. The propulsion of sea turtles has been shown to occur for approximately 30% of the limb beat, with the remaining 70% employing a drag-reducing glide. However, it is unknown how the sea turtle manipulates the flow during the propulsive stage. Answering this research question is a complicated process, especially when conducting laboratory tests on endangered animals, and the animal may not even swim with its regular routine while in a captive state. In this work, we take advantage of our robotic sea turtle, internally known as Cornelia, to offer the first insights into the flow features during the sea turtle’s propulsion cycle consisting of the downstroke and the sweep stroke. Comparing the flow features to the animal’s swim speed, flipper angle of attack, power consumption, thrust and lift production, we hypothesise how each of the flow features influences the animal’s propulsive efforts and cost of transport (COT). Our findings show that the sea turtle can produce extremely low COT values that point to the effectiveness of the sea turtle propulsive technique. Based on our findings, we extract valuable data that can potentially lead to turtle-inspired elements for high-efficiency underwater drones for long-term underwater missions.

This paper investigates the implementation of a novel evolvable hardware controller used in evolutionary robotics. The evolvable hardware consists of a Cartesian based array of logic blocks comprised of multiplexers and logic elements. The logic blocks are configured by a bit stream which is evolved using a genetic algorithm. A comparison is performed between an evolvable hardware and an artificial neural network controller evolved using the same genetic algorithm to produce the gait of a hexapod robot. To compare the two controllers, differences in their evolutionary efficiency and robot performance are investigated. The evolutionary efficiency is measured by the required number of generations to achieve an optimal fitness. An optimal hexapod controller allows the robot to walk forward in a straight line maintaining a constant heading and body attitude. It was found that the evolutionary efficiency and performance of the evolvable hardware and artificial neural network were similar, however the EHW was the most evolutionary efficient requiring less generations on average to evolve. Both evolved controllers were evaluated in simulation, and on a physical robot using a softcore processor and custom hardware implemented on a FPGA. The implementation showed that the controllers performed equally well when deployed, allowing the hexapod to meet the optimal gait criteria. These findings have shown that the evolvable hardware controller is a valid option for robotic control of a multilegged robot such as a hexapod as its evolutionary efficiency and deployed performance on a real robot is comparable to that of an artificial neural network. One future application of these evolvable controllers is in fault tolerance where the robot can dynamically adapt to a fault by evolving the controller to adjust to the fault conditions.