Closeup exploration of underwater life requires new forms of interaction, using biomimetic creatures that are capable of agile swimming maneuvers, equipped with cameras, and supported by remote human operation. Current robotic prototypes do not provide adequate platforms for studying marine life in their natural habitats. This work presents the design, fabrication, control, and oceanic testing of a soft robotic fish that can swim in three dimensions to continuously record the aquatic life it is following or engaging. Using a miniaturized acoustic communication module, a diver can direct the fish by sending commands such as speed, turning angle, and dynamic vertical diving. This work builds on previous generations of robotic fish that were restricted to one plane in shallow water and lacked remote control. Experimental results gathered from tests along coral reefs in the Pacific Ocean show that the robotic fish can successfully navigate around aquatic life at depths ranging from 0 to 18 meters. Furthermore, our robotic fish exhibits a lifelike undulating tail motion enabled by a soft robotic actuator design that can potentially facilitate a more natural integration into the ocean environment. We believe that our study advances beyond what is currently achievable using traditional thruster-based and tethered autonomous underwater vehicles, demonstrating methods that can be used in the future for studying the interactions of aquatic life and ocean dynamics.

Exploration of underwater life with an acoustically controlled soft robotic fish

Generating grasp poses is a crucial component for any robot object manipulation task. In this work, we formulate the problem of grasp generation as sampling a set of grasps using a variational autoencoder and assess and refine the sampled grasps using a grasp evaluator model. Both Grasp Sampler and Grasp Refinement networks take 3D point clouds observed by a depth camera as input. We evaluate our approach in simulation and real-world robot experiments. Our approach achieves 88% success rate on various commonly used objects with diverse appearances, scales, and weights. Our model is trained purely in simulation and works in the real-world without any extra steps.

/pdf/6-dof-graspnet-variational-grasp-generation-for-object-42adjti827.pdf

6-DOF GraspNet: Variational Grasp Generation for Object Manipulation

Gelatinous underwater invertebrates such as jellyfish have organs that are transparent, stretchable, touch-sensitive and self-healing, which allow the creatures to navigate, camouflage themselves and, indeed, survive in aquatic environments. Artificial skins that emulate such functionality could be used to develop applications such as aquatic soft robots and water-resistant human–machine interfaces. Here we report a bio-inspired skin-like material that is transparent, electrically conductive and can autonomously self-heal in both dry and wet conditions. The material, which is composed of a fluorocarbon elastomer and a fluorine-rich ionic liquid, has an ionic conductivity that can be tuned to as high as 10−3 S cm−1 and can withstand strains as high as 2,000%. Owing to ion–dipole interactions, it offers fast and repeatable electro-mechanical self-healing in wet, acidic and alkali environments. To illustrate the potential applications of the approach, we used our electronic skins to create touch, pressure and strain sensors. We also show that the material can be printed into soft and pliable ionic circuit boards. A transparent electronic skin, composed of an elastomer and an ionic liquid, can autonomously self-heal in both dry and wet conditions due to ion–dipole interactions.

Self-healing electronic skins for aquatic environments

IEEE International Conference on Robotics and Automation (ICRA) におけるフルードパワー技術の研究動向

Origami robots are created using folding processes, which provide a simple approach to fabricating a wide range of robot morphologies. Inspired by biological systems, engineers have started to explore origami folding in combination with smart material actuators to enable intrinsic actuation as a means to decouple design from fabrication complexity. The built-in crease structure of origami bodies has the potential to yield compliance and exhibit many soft body properties. Conventional fabrication of robots is generally a bottom-up assembly process with multiple low-level steps for creating subsystems that include manual operations and often multiple iterations. By contrast, natural systems achieve elegant designs and complex functionalities using top-down parallel transformation approaches such as folding. Folding in nature creates a wide spectrum of complex morpho-functional structures such as proteins and intestines and enables the development of structures such as flowers, leaves and insect wings. Inspired by nature, engineers have started to explore folding powered by embedded smart material actuators to create origami robots. The design and fabrication of origami robots exploits top-down, parallel transformation approaches to achieve elegant designs and complex functionalities. In this Review, we first introduce the concept of origami robotics and then highlight advances in design principles, fabrication methods, actuation, smart materials and control algorithms. Applications of origami robots for a variety of devices are investigated, and future directions of the field are discussed, examining both challenges and opportunities. Inspired by biological systems, engineers are exploring origami folding with smart material actuation to enable intrinsically actuated designs with complex functionalities and easy fabrication. This Review highlights recent advances in the design, fabrication and control of these origami robots.

Design, fabrication and control of origami robots

Communication with a robot using brain activity from a human collaborator could provide a direct and fast feedback loop that is easy and natural for the human, thereby enabling a wide variety of intuitive interaction tasks. This paper explores the application of EEG-measured error-related potentials (ErrPs) to closed-loop robotic control. ErrP signals are particularly useful for robotics tasks because they are naturally occurring within the brain in response to an unexpected error. We decode ErrP signals from a human operator in real time to control a Rethink Robotics Baxter robot during a binary object selection task. We also show that utilizing a secondary interactive error-related potential signal generated during this closed-loop robot task can greatly improve classification performance, suggesting new ways in which robots can acquire human feedback. The design and implementation of the complete system is described, and results are presented for realtime closed-loop and open-loop experiments as well as offline analysis of both primary and secondary ErrP signals. These experiments are performed using general population subjects that have not been trained or screened. This work thereby demonstrates the potential for EEG-based feedback methods to facilitate seamless robotic control, and moves closer towards the goal of real-time intuitive interaction.

Correcting robot mistakes in real time using EEG signals

We present a three-dimensional deep convolutional neural network (3D CNN) approach for grasping unknown objects with soft hands. Soft hands are compliant and capable of handling uncertainty in sensing and actuation, but come at the cost of unpredictable deformation of the soft fingers. Traditional model-driven grasping approaches, which assume known models for objects, robot hands, and stable grasps with expected contacts, are inapplicable to such soft hands, since predicting contact points between objects and soft hands is not straightforward. Our solution adopts a deep CNN approach to find good caging grasps for previously unseen objects by learning effective features and a classifier from point cloud data. Unlike recent CNN models applied to robotic grasping which have been trained on 2D or 2.5D images and limited to a fixed top grasping direction, we exploit the power of a 3D CNN model to estimate suitable grasp poses from multiple grasping directions (top and side directions) and wrist orientations, which has great potential for geometry-related robotic tasks. Our soft hands guided by the 3D CNN algorithm show 87% successful grasping on previously unseen objects. A set of comparative evaluations shows the robustness of our approach with respect to noise and occlusions.

Learning Object Grasping for Soft Robot Hands

This paper describes a novel underactuated robotic hand design. The hand is highly underactuated as it contains three fingers with three joints each controlled by a single motor. One of the fingers (“thumb”) can also be rotated about the base of the hand, yielding a total of two controllable degrees-of-freedom. A key component of the design is the addition of position and tactile sensors which provide precise angle feedback and binary force feedback. Our mechanical design can be analyzed theoretically to predict contact forces as well as hand position given a particular object shape

/pdf/a-highly-underactuated-robotic-hand-with-force-and-joint-2o2jjsrt6a.pdf

A highly-underactuated robotic hand with force and joint angle sensors

Seamless communication of desired motions and goals is essential for enabling effective physical human-robot collaboration. In such cases, muscle activity measured via surface electromyography (EMG) can provide insight into a person’s intentions while minimally distracting from the task. The presented system uses two muscle signals to create a control framework for team lifting tasks in which a human and robot lift an object together. A continuous setpoint algorithm uses biceps activity to estimate changes in the user’s hand height, and also allows the user to explicitly adjust the robot by stiffening or relaxing their arm. In addition to this pipeline, a neural network trained only on previous users classifies biceps and triceps activity to detect up or down gestures on a rolling basis; this enables finer control over the robot and expands the feasible workspace. The resulting system is evaluated by 10 untrained subjects performing a variety of team lifting and assembly tasks with rigid and flexible objects.

Joseph DelPreto

Papers

Exploration of underwater life with an acoustically controlled soft robotic fish

Correcting robot mistakes in real time using EEG signals

Learning Object Grasping for Soft Robot Hands

A highly-underactuated robotic hand with force and joint angle sensors

Sharing the Load: Human-Robot Team Lifting Using Muscle Activity