Mobile UnderSEA HAND for whole hand robust grasping – SEAHAND

The starting point for the development process of a gripper capable of replacing the archaeologist's hand is the analysis of the expert's gestures with characteristic objects, such as those collected in a marine environment. The recorded data constitute the entry point for the gripper synthesis process (kinematics, sizing, torques and joint speeds). To allow equal pressure of the robotic hand and guarantee its watertightness despite the abrasive environment present in the seabed, a specific flexible glove was developed, filled with an incompressible fluid. The evaluation of the forces applied by the gripper's fingers is essential to be able to safely grasp objects at great depths given the poor visibility. The high pressure of the seabed makes the use of conventional external force sensors impossible, which led to the proposal of an original method for measuring clamping forces, based on the evaluation of the elongation of the cables that actuate the joints. In terms of piloting, the human hand of the archaeologist operator is used as a reference to control the SEAHAND robotic hand from the surface boat and also the underwater vehicle carrying the hand, which has been the subject of the development of an original control. The visual feedback of the forces perceived by each finger is thus combined with a remote-controlled control of the gripper and the submarine.

To meet the challenges of the SEAHAND project, and in order to validate the design of the gripper's cable transmission and force evaluation technology upstream, the project team relied on improvements to the “terrestrial” RoBioSS robotic hand, on which the force evaluation technology was qualified upstream, while improving the robustness of the cable transmission. With the support of the SEAHAND project, improvements to the RoBioSS hand were validated and implemented on the SEAHAND gripper. The robotic gripper research activity has thus been the subject of significant communication and valorisation activity based on the project's developments from 2015 to 2021. This original prototype of an underwater gripper, comprising 4 fingers and 6 actuators, is designed to pick up fragile objects in the context of underwater archaeology. Over the years, the project team has carried out extensive collaborative work involving all partners, with the aim of being able to experiment with the gripper in the water at the end of the project. Two CNRS patents have been issued for the gripper developed. In 2023, the gripper was put to the test on the nacelle of a parallel robot with a large terrestrial cable to evaluate complex adaptive gripping scenarios, foreshadowing a future application on a mini submarine.

The SEAHAND project opens up great prospects for development in 2021 to assist humans in remote handling tasks in a marine environment. The complementarity of the partners and the strong involvement of each has offered an ideal partnership dynamic that we wish to continue beyond the project. The coming months will allow us to evaluate the gripper in real configuration with the support of the new DRASSM vessel, inaugurated in July 2021, and we hope to continue developments in this context following these tests. The work carried out already allows us to envisage great scientific prospects to support complex manipulations that are not feasible today. We also thank TECHNO CONCEPT for its support in marine design engineering, COURBIS for the development of the specific glove and CATIE for the development of the on-board electronics.

Creuze, V. Monocular Odometry for Underwater Vehicles with Online Estimation of the Scale Factor. IFAC 2017 World Congress, 20th World Congress of the International Federation of Automatic Control. Toulouse, France, July 9-14, 2017.

Fischer, H.; Vulliez, M.; Laguillaumie, P.; Vulliez, P.; Gazeau, J.-P. A framework for real-time multi-axis and multi-robot control; IEEE Transactions on Automation Science and Engineering. July 2019, 16, 3.

Mizera, C.; Delrieu, T.; Weistroffer, V.; Andriot, C.; Decatoire, A.; Gazeau, J.-P. Evaluation of hand-tracking systems in teleoperation and virtual dexterous manipulation. IEEE Sensors Journal. Feb 2020, 20, 3.

Mnyussiwalla, H.; Seguin, P.; Vulliez, P.; Gazeau, J.-P. Evaluation and selection of grasp quality criteria for dexterous manipulation. Journal of Intelligent & Robotic Systems. 2022, 104, 2, 1-31.

Gazeau, J.-P.; Laguillaumie, P.; Vulliez, P.; Mizera, C. Robotic hand sensitive to forces in an aquatic environment. 17757351. 2/2/2023.

The SEAHAND project aims at developing a new paradigm of manipulator for the excavation of deep sea archaeological sites or shipwrecks (up to 2000 meters deep). In fact, we propose a mobile adaptive grasper, carried by a very small fully actuated underwater vehicle. This approach will allow remote manipulation of delicate objects, providing force feedback along all degrees of freedom (including the squeezing force along each finger). The design of the hand and the vehicle will be conducted together by the involved partners, taking into account the constraints impose by the force-feedback control approach. The result will be an innovative mobile grasper allowing archaeologists to handle very fragile artefacts under very high water pressure (200 bars). The force feedback will allow safe excavation of the objects and will contribute to preserve their integrity.
The joint action of specialists of all the concerned fields (design of underwater vehicles and robotics hands, remote manipulation in nuclear environment, control), associated with the expertise of world renowned underwater archaeologists, will allow the development of a relevant industrial prototype, that will be validated during real field tests.