A new aquatic robot inspired by Manta rays has broken the world record for the fastest swimming soft robot. The robot, designed by a team of engineers from North Carolina State University and the University of Virginia, was able to reach speeds of 6.8 body lengths per second. That comes out to a swim speed of 156.4 mm per second or about 0.35 mph. That time blows past the previous record of 3.74 body lengths per second record previously set by the same researchers. Researchers behind the machine, who published their findings today in Science Advances, told Popular Science the new design could be useful for future deep-sea exploration efforts.

Underwater vehicles have seen significant development over the past seventy years. However, bio-inspired propulsion robots are still in their early stages and require greater interdisciplinary collaboration between biologists and roboticists. The octopus, one of the most intelligent marine animals, exhibits remarkable abilities such as camouflaging, exploring, and hunting while swimming with its arms. Although bio-inspired robotics researchers have aimed to replicate these abilities, the complexity of designing an eight-arm bionic swimming platform has posed challenges from the beginning. In this work, we propose a novel bionic robot swimming platform that combines asymmetric passive morphing arms with an umbrella-like quick-return mechanism. Using only two simple constant-speed motors, this design achieves efficient swimming by replicating octopus-like arm movements and stroke time ratios. The robot reached a peak speed of 314 mm/s during its second power stroke. This design reduces the complexity of traditional octopus-like swimming robot actuation systems while maintaining good swimming performance. It offers a more achievable and efficient platform for biologists and roboticists conducting more profound octopus-inspired robotic and biological studies.

Exploring bodies of water on their surface allows robots to efficiently communicate and harvest energy from the sun. On the water surface, however, robots often face highly unstructured environments, cluttered with plant matter, animals, and debris. We report a fast (5.1 cm/s translation and 195 {\deg}/s rotation), centimeter-scale swimming robot with high maneuverability and autonomous untethered operation. Locomotion is enabled by a pair of soft, millimeter-thin, undulating pectoral fins, in which traveling waves are electrically excited to generate propulsion. The robots navigate through narrow spaces, through grassy plants, and push objects weighing over 16x their body weight. Such robots can allow distributed environmental monitoring as well as continuous measurement of plant and water parameters for aqua-farming.

In this paper, we present an exploration and assessment of employing a centralized deep Q-network (DQN) controller as a substitute for the prevalent use of PID controllers in the context of 6DOF swimming robots. Our primary focus centers on illustrating this transition with the specific case of underwater object tracking. DQN offers advantages such as data efficiency and off-policy learning, while remaining simpler to implement than other reinforcement learning methods. Given the absence of a dynamic model for our robot, we propose an RL agent to control this multi-input-multi-output (MIMO) system, where a centralized controller may offer more robust control than distinct PIDs. Our approach involves initially using classical controllers for safe exploration, then gradually shifting to DQN to take full control of the robot. We divide the underwater tracking task into vision and control modules. We use established methods for vision-based tracking and introduce a centralized DQN controller. By transmitting bounding box data from the vision module to the control module, we enable adaptation to various objects and effortless vision system replacement. Furthermore, dealing with low-dimensional data facilitates cost-effective online learning for the controller. Our experiments, conducted within a Unity-based simulator, validate the effectiveness of a centralized RL agent over separated PID controllers, showcasing the applicability of our framework for training the underwater RL agent and improved performance compared to traditional control methods. The code for both real and simulation implementations is at https://github.com/FARAZLOTFI/underwater-object-tracking.

Soft robots have a myriad of potentials because of their intrinsically compliant bodies, enabling safe interactions with humans and adaptability to unpredictable environments. However, most of them have limited actuation speeds, require complex control systems, and lack sensing capabilities. To address these challenges, here we geometrically design a class of metacaps whose rich nonlinear mechanical behaviors can be harnessed to create soft robots with unprecedented functionalities. Specifically, we demonstrate a sensor-less metacap gripper that can grasp objects in 3.75 ms upon physical contact and a pneumatically actuated gripper with tunable actuation behaviors that have little dependence on the rate of input. Both grippers can be readily integrated into a robotic platform for practical applications. Furthermore, we demonstrate that the metacap enables propelling of a swimming robot, exhibiting amplified swimming speed as well as untethered, electronics-free swimming with tunable speeds. Our metacaps provide new strategies to design the next-generation soft robots that require high transient output energy and are capable of autonomous and electronics-free maneuvering.

Researchers have designed miniature robots that are inspired by cells and steered by ultrasound that could one day navigate the human body and help deliver drugs to certain parts of it. These'rocket ships,' as described by scientists at Cornell University, have a design that is inspired by both bacteria and sperm cells. The robots, which could navigate through the human body are controlled remotely and could take advantage of some features of sperm and bacteria cells, including the fact that bacteria can swim 10 times their body length and sperm can go against the flow. 'We can make airplanes that are better than birds nowadays,' said study co-author, Mingming Wu, professor of biological and environmental engineering at Cornell, in a statement. 'But at the smallest scale, there are many situations that nature is doing much better than us.

Scientists at the Biorobotics Laboratory (BioRob) in EPFL's School of Engineering are developing innovative robots in order to study locomotion in animals and, ultimately, gain a better understanding of the neuroscience behind the generation of movement. One such robot is AgnathaX, a swimming robot employed in an international study with researchers from EPFL as well as Tohoku University in Japan, Institut Mines-Te le com Atlantique in Nantes, France, and Universite de Sherbrooke in Canada. The study has just been published in Science Robotics. "Our goal with this robot was to examine how the nervous system processes sensory information so as to produce a given kind of movement," says Prof. Auke Ijspeert, the head of BioRob and a member of the Rescue Robotics Grand Challenge at NCCR Robotics. "This mechanism is hard to study in living organisms because the different components of the central and peripheral nervous systems* are highly interconnected within the spinal cord. That makes it hard to understand their dynamics and the influence they have on each other."

As the world contemplates ways to deploy robots to help with hazardous tasks such as large-scale cleanups or search and rescue missions, scientists are also second-guessing how to fix machines if they come into harm's way. We've previously seen jelly-like polymers used to create robotic hands that can repair themselves after a violent infliction. And, "biological" androids made from stem cells that can regenerate and stitch back together when sliced. The latest breakthrough involves tiny microbots that can magnetically "heal" themselves on the fly after breaking apart, without help from humans. Researchers at UC San Diego did this by creating 2 cm long swimming robots shaped like fish and composed of three layers.

So it looks like a half-stuffed sock -- and it is, sort of -- but this sandfish-inspired search and rescue robot has the potential to change the way machines maneuver through disaster zones. Playing off its previous endeavors, a team of Georgia Tech researchers has designed a wedge-shaped head to manipulate the vertical movement of its sand-swimming invention through "complex dirt and rubble environments." By mimicking the pointy snout of the sandfish lizard, and attaching it to the body of its robot -- which sports seven servo-powered segments stuffed in a latex sock and sheathed by a spandex "swimsuit" -- the team found that subtle changes in the positioning of the robot's head made for drastic differences in vertical movement. When it was placed flat on the horizontal plane, the robot descended; when it was inclined above seven degrees, it ascended. For now, the robotic sandfish has been relegated to swimming in a sea of tiny yellow balls, but it's slated to dive into a pool of debris in the name of research soon.

For those who don't have the patience to wait for fish to take the bait, a robotics firm has unveiled a device that finds and attracts them to you. Called the PowerRay, the underwater drone uses a precision remote bait drop and blue-hued fishing lures that basically attaches the fish to the dangling hook. The unmanned vehicle also transmits data, fish alerts and live video feed and photos of the underwater world to the fisherman's smartphone via a companion app. PowerRay uses a precision remote bait drop and blue-hued fishing lures that basically attaches the fish to the dangling hook. The unmanned vehicle also transmits data, alters and live video feed and photos of the underwater world to the fisherman's smartphone via a companion app The sonar system can detect fish up to 131ft below the swimming robot and can dive up to 98 feet.