RE2 Robotics, a leading developer of intelligent mobile manipulation systems, has recently been awarded a contract by the Office of Naval Research to create an autonomous robotic system to subdue underwater mines. This system will require integrating multiple artificial intelligence techniques, ranging from computer vision to machine autonomy. The goal of this project is to not only remove these mines from the ocean but also provide a safer underwater environment.
Image Credit: archy13/Shutterstock.com
Naval mines are enclosed explosive devices placed in the water to destroy either submarines or surface ships. Their use was widespread throughout the First and Second World Wars.
Since 1988, countries producing mines have increased by 75%, with many also having the capability to deploy these mines strategically. Despite international law requiring states to disclose areas where mines have been deployed, their specific locations in the ocean remain state secrets. Further to this, numerous states do not comply with these laws, increasing the threat of unknown mines in the oceans.
To combat this, the U.S Navy has awarded a contract valued at $9.5 million to intelligent robotics company RE2, tasking them with the development of systems that can locate and clear these mines. The project, named the Maritime Mine Neutralization System (M2NS), will integrate several RE2 systems: RE2 Sapien Sea Class, RE2 Detect, and RE2 Intellect, into a single system built for localization and clearing of naval mines.
The systems constructed by RE2 will be designed to attach neutralization devices to mines once they are found and activate them.
RE2 Sapien Sea Class
The RE2 Sapien is a family of robotic arms, of which the Sea Class variant is the most suitable for this project. These robotic manipulators are designed to provide human-like control while operating underwater, as the systems consist of two six-degree-of-freedom arms with three-fingered end effectors.
The RE2 Sapien Sea Class also has various control capabilities; it can be teleoperated or operated through autonomous or semi-autonomous systems.
This teleoperation is conducted through the RE2 Imitative Controller, a device that a user can wear to control the manipulators aboard the RE2 Sapien Sea Class. For autonomous and semi-autonomous control, various perceptual systems are used.
Alongside this, its neutral buoyancy system also allows the system to be operated at any depth. Combining these features means that the RE2 Sapien Sea Class provides the ideal underwater solution for the M2NS project.
RE2 Detect
For any machine autonomy, the machine needs to continuously acquire information about the environment it is operating in and utilize it in a way that best achieves its current goal.
To do this, robotic systems can employ a multitude of cameras and sensors to gather data from the environment. This data can be anything from visual data in the form of images to spatial data in the form of point clouds from lidar sensors.
RE2 Sapien arms contain a multi-modal system consisting of 2D and 3D imaging sensors to retrieve environmental information. Alongside this, RE2 has developed its own proprietary algorithms that allow object detection within any indoor or outdoor environment.
Such a development is significant, as most systems are built using very select data, meaning that they are restricted from operating in certain environments or offer lower performance when working outside of these trained scenarios.
Having a system such as this provides a perfect component to be integrated into autonomous machinery.
MDMS: Autonomous Underwater Manipulation
Video Credit: RE2, Inc./YouTube.com
RE2 Intellect
To have a truly intelligent machine, there must be some form of decision-making based on the machine's data and their desired goal. RE2 Intellect is a system that has been designed to work alongside the RE2 Detect systems to endow these robotic systems with human-like decision-making.
The RE2 Intellect algorithms combine the 2D and 3D computer vision capabilities with machine learning and deep learning algorithms to guide the control of the robot. The software developed by RE2 is useful in structured environments and is also robust to anomalies in the environment, and can adapt to environmental changes.
As such, the RE2 Intellect software presents an ideal way to control an autonomous system in an underwater scenario. With the vast majority of the ocean unexplored, the ability to adapt in a goal-directed manner will prove invaluable for navigating under the ocean.
Future Outlooks
The M2NS project is an ambitious one, not only for the sector of robotics but also for marine safety and conservation. The benefits of this project will not only be seen by its funder, the Office of Naval Research, but also by many other private individuals. This includes fishing companies, shipping providers, and ocean transport providers; once cleared, new ocean areas will become accessible and allow for an increase in viable ocean routes.
As time goes on, we will also see underwater robotics technology increase, with projects like this providing much of the groundwork to build on top of.
Continue reading: Unmanned Aerial Vehicles in the Australian Defense Industry.
References and Further Reading
RE2 Robotics (2020) RE2 Robotics Receives $9.5 Million Contract to Develop Underwater Autonomous System for the U.S. Navy. [Online] Available at: https://www.resquared.com/blog/re2-to-develop-underwater-autonomous-system
RE2 Robotics (2020) RE2 Sea Class. [Online] Available at: https://www.resquared.com/underwater
RE2 Robotics (2020) RE2 Products. [Online] Available at: https://www.resquared.com/products
Vasilev, V (2018) International regulations related to naval mines and protection of navigation, IOP Conf. Ser.: Earth Environ. Sci. 172 012013. Available at: https://doi.org/10.1088/1755-1315/172/1/012013
Vasilev, V., 2018. International regulations related to naval mines and protection of navigation. IOP Conference Series: Earth and Environmental Science, 172, p.012013.
Disclaimer: The views expressed here are those of the author expressed in their private capacity and do not necessarily represent the views of AZoM.com Limited T/A AZoNetwork the owner and operator of this website. This disclaimer forms part of the Terms and conditions of use of this website.