Waterbug - An autonomous mm-scale robot on the water surface (white paper tracking # 22-000003438)
Abstract
The water surface plays a critical role in the environment, providing habitats for aquatic life, and supporting sources for drinking water, hydropower, and irrigation. Autonomous robots that can operate in these areas have a number of applications from sensing and tracing micro-elements or pollutants, monitoring aquatic life, and detecting human activity. Smaller, cm-scale robots have a number of advantages over larger boat- or glider vessels: they can (1) survey and operate inconspicuously in a range of waters, (2) pose less of an environmental impact in the case of partial failure, and (3) are faster and less expensive to produce, thereby supportingfuture swarms of robots, which can provide adaptability and resilience. As the characteristic dimension of a device decreases, surface forces begin to dominate relative to Newtonian forces, allowing for unique opportunities to generate momentum. Insects like water striders and fishing spiders can remain on the water#s surface without active locomotion strategies, utilizing surface tension to maintain contact with the interface. To maneuver along the water surface, these insects generate surface waves, using active legs, hydrofoils, or flapping wings, to impart momentum to the water. To address the challenges associated with generating autonomous locomotion on the air-water interface, we will take inspiration from natural surface dwellers. Specifically, we propose to develop an autonomous, cm-scale robotic platform that operates on the air-water interface, focusing on mechanism design, power, and control. We will optimize locomotion modalities on the water surface, utilizing a combination of flapping wings and legged locomotion strategies. Leveraging the significant progress we have made in the design and manufacturing of cm-scale mechanisms, we will build new robotic platforms and characterize their performance on the water surface, quantifying tradeoffs in force generation, velocity, power density, and cost of transport. Using biological analogs, we will develop fundamental understanding of locomotion on the air-water interface to enhance maneuverability, enabling rapid turns and sudden stops. Our first goal is to demonstrate a cm-scale (<50mm), low mass (<500mg) robot that is capable of maneuvering across the 2D interface. Because these robots need to operate in unstructured environments, we need to develop an embedded system that enables controlled, power autonomous locomotion. First, the controller will need both proprioceptive and exteroceptive feedback to estimate the robot#s state and changes in the surrounding environment. When available, we will integrate commercial off the shelf (COTS) components that meet the size, weight, and power requirements of the vehicle. Wewill also develop custom analog sensing strategies (i.e., concomitant sensing of the actuators, capacitive elements in the legs, and bio-inspired sun compasses) that can be embedded into the robotic platform to reduce overall system mass. Second, we will design high efficiency power electronics systems coupled with small-scale power sources to enable power autonomous locomotion. We will benchmark computation requirements for these sensing and power systems, and integrate COTS microcontroller units to compute control signals for autonomous heading control. This work will lay the foundation for a new class of cm-scale robots that can operate autonomously in an aquatic environment. This work will create a resilient platform for future research directions: (1) the design of low-power communication systems and coordination strategies of multiagent systems, (2) estimation and control policies for task-specific deployments, (3) mm-scale exteroceptive sensing suites to detect and trace elements on the water surface, (4) efficient compute for severe resource-constrained robots, and (5) multimodal locomotion capabilities to effectively transition from the surface or water to ground.Approved for public release.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Mar 08, 2024
- Source ID
- N000142412141
Entities
People
- E Helbling
Organizations
- Cornell University
- Office of Naval Research
- United States Navy