The SERDP and ESTCP Munitions Response Program held its spring In-Progress Review virtually last week. We wanted to highlight two projects focused on unique underwater munitions response issues.

For autonomous robots, big waves and shoreline terrain can be big problems for small robots. Waves can overturn small robots, so typically, the larger the waves the larger the robot needs to be. At the same time, sand can trap robots and rocks can be too large for robots to cross. This limits autonomous robots that could be valuable tools for munitions response. In contrast, biological crabs at a range of sizes thrive in shallow water.

In her SERDP project, Dr. Kathryn Daltorio of Case Western Reserve University hypothesizes that an essential part of a crab’s strategy is that they use waves and terrain against each other. If their legs can grasp the terrain, then they can resist the hydrodynamic wave forces. At the same time, if they can release their grip on the terrain at just the right time, they might be able to overcome obstacles that are otherwise too big. Such a gait might be more power efficient as well.

To test this, Dr. Daltorio’s team is building and testing amphibious legged robots. They have found that gripping the ground with crab-like dactyls increases anchoring, even in sand, such that the effective weight is increased by approximately ⅓. To test this with consistent waves, they have utilized a wave tank in their lab to show that dactyls reduce travel from waves.

As a legacy of years of military activities, including training and testing, there are numerous current and former Department of Defense aquatic sites contaminated with munitions and explosives of concern. At many inland waters and coastal areas, significant efforts have been mandated to manage and clean-up these sites to reduce the risk of human interaction with unexploded ordnance (UXO). Effective management requires knowing where the UXO are located and whether they are buried in the seafloor sediment. In this SERDP project, Dr. Sarah Rennie, Dr. Alan Brandt, and other researchers at Johns Hopkins University Applied Physics Laboratory (JHU/APL) have been developing a probabilistic expert system to predict underwater migration potential and depth of burial of munitions residing in the seabed.

The framework of the Underwater Munitions Expert System (UnMES) is based on a Bayesian Network (BN) construct, a useful method of modeling systems in a probabilistic manner. UnMES relates the extant hydrodynamic and geological site characteristics with the response of different munitions. Results from multiple SERDP MR projects provide the basis of the process models used in the system.

The Underwater Munitions Expert System is designed to be a module in a comprehensive computer-based decision support tool to assist planning and decision-making in site remediation management.