It has been a great challenge to develop robots that are able to perform complex movement patterns with high speed and, simultaneously, high accuracy. Copepods are animals found in freshwater and saltwater habitats that can have extremely fast escape responses when a predator is sensed by performing explosive curved jumps. Here, we present a design and build prototypes of a combustion-driven underwater soft robot, the "copebot", that, like copepods, is able to accurately reach nearby predefined locations in space within a single curved jump. Because of an improved thrust force transmission unit, causing a large initial acceleration peak (850 Bodylength*s-2), the copebot is 8 times faster than previous combustion-driven underwater soft robots, whilst able to perform a complete 360{\deg} rotation during the jump. Thrusts generated by the copebot are tested to quantitatively determine the actuation performance, and parametric studies are conducted to investigate the sensitivities of the input parameters to the kinematic performance of the copebot. We demonstrate the utility of our design by building a prototype that rapidly jumps out of the water, accurately lands on its feet on a small platform, wirelessly transmits data, and jumps back into the water. Our copebot design opens the way toward high-performance biomimetic robots for multifunctional applications.

翻译：开发能够以高速同时实现高精度复杂运动模式的机器人一直是一项巨大挑战。桡足类是一类在淡水和咸水栖息地中发现的动物，当感知到捕食者时，它们能通过爆发性曲线跳跃执行极快的逃逸反应。本文提出了一种燃烧驱动的水下软体机器人"copebot"的设计方案并构建了原型机，该机器人如同桡足类动物一样，能够通过单次曲线跳跃精确到达空间中邻近的预定位置。由于改进的推力传递单元产生了较大的初始加速度峰值（850倍体长/秒²），copebot的速度是此前燃烧驱动水下软体机器人的8倍，同时能在跳跃过程中完成完整的360°旋转。通过测试copebot产生的推力以定量评估其驱动性能，并开展参数化研究探究输入参数对其运动学性能的敏感度。我们通过构建原型机验证了设计的实用性：该机器人能够快速跃出水面、精准降落在小型平台上四足着陆、无线传输数据并跳回水中。我们的copebot设计为多功能的仿生高性能机器人开辟了道路。