Experiments with Cooperative Control of Underwater Robots

Matthew Dunbabin, Peter Corke, Iuliu Vasilescu, Daniela Rus
2009 The international journal of robotics research  
In this paper we describe cooperative control algorithms for robots and sensor nodes in an underwater environment. Cooperative navigation is defined as the ability of a coupled system of autonomous robots to pool their resources to achieve long-distance navigation and a larger controllability space. Other types of useful cooperation in underwater environments include: exchange of information such as data download and retasking; cooperative localization and tracking; and physical connection
more » ... ing) for tasks such as deployment of underwater sensor networks, collection of nodes, and rescue of damaged robots. We present experimental results obtained with an underwater system that consists of two very different robots and a number of sensor network modules. We present the hardware and software architecture of this underwater system. We then describe various interactions between the robots and sensor nodes and between the two robots, including cooperative navigation. Finally, we describe our experiments with this underwater system and present data. (2) the cooperative navigation algorithm, (3) the application of this algorithm to data muling, and (4) the experimental characterization of cooperative navigation. Cooperation is an important aspect of designing useful underwater robots. Long-range and long-endurance underwater operations require substantial power and thus a large robot. Once at the destination the large size may be a disadvantage if the task requires high maneuverability or a small size, for example, to enter a wreck. We would like to have robot systems that meet both needs: (1) they can perform long-range travel while also (2) being maneuverable at the destination. We propose cooperation as a means for accomplishing this. One approach to realizing these two goals is to create heterogeneous robots that will cooperate in long distance travel by docking together and taking advantage of the increased number of resources (e.g. thrusters and power) in the system. Once at the destination, they can disassemble into, or unload, smaller and more maneuverable and perhaps task-specific robots that work individually or in concert. Many tasks require localization and this could be achieved by a cooperative system of robots that self-deploy in a way that forms a networked system for high-precision distributed localization and tracking [Moore et al., 2004 ,Detweiler et al., 2006 . Finally, such cooperating systems could sustain long-term operation by designating one robot as a power and recharging station. An alternative approach is modular cooperative underwater robots that can deploy and recover sensor networks. The deployer is an autonomous vehicle with stackable sensing modules of uniform shape. Each module has a computation, motor, buoyancy, and battery module, and an arbitrary number of sensor modules. A docking mechanism allows a module to attach to the one above it in the structure. A fleet of such robots can be controlled to place sensors at designated locations on pipelines, pillars, or specific locations at the bottom of the ocean. Once in place, each sensor collects data, assists with monitoring tasks, and participates in guiding the AUVs. The robots can also be used to reposition the network to repair and maintain network connectivity (similar to the ideas in [Corke et al., 2004a , Corke et al., 2004b ). The multi-module nature of the system coupled with its ability to dock to arbitrary units also provides redundancy: when a robot is damaged, a different robot can locate it, dock with it, and retrieve it. More specifically, in this paper we build on our work in [Dunbabin et al., 2006b ] and describe our first steps toward the vision of cooperative modular underwater robots. We present a heterogeneous system that consists of two robots with very different capabilities and a collection of static underwater sensor nodes, and all system elements are networked acoustically and optically. When coupled together, the robots can travel further, faster, and along a larger set of trajectories than each individual robot. Because the robots have different types of perception and communication resources, the coupled system can also execute more complex tasks than either robot alone. We focus on the algorithmic and systems issues related to enabling these robots to become physically coupled and to coordinate navigation by sharing resources (thrusters and sensors). A docking mechanism and algorithm enables the two robots to become physically coupled. We describe in detail how we achieve navigation control and present experimental data from several trials in a test tank. We describe how one robot is able to rescue another, and finally, we describe an application of the cooperative system to data muling from an in-situ sensor network. Much work remains to be done to achieve the full potential of cooperative underwater robotics. We have encouraging initial steps in this direction.
doi:10.1177/0278364908098456 fatcat:bg7mib35wfbazojpy4ehobpwci