Overview

As part of the NAMOS project we have designed and developed a sensor-actuated network for aquatic monitoring. The network consists of ten stationary buoys and one mobile robotic boat for real-time, in-situ measurements and analysis of chemical and physical factors governing the abundances and dynamics of microorganisms at biologically-relevant spatiotemporal scales. The goal of the network is to obtain high-resolution information on the spatial and temporal distributions of plankton assemblages and concomitant environmental parameters in aquatic environments using the in-situ presence afforded by the network, and to make possible network-enabled robotic sampling of hydrographic features of interest. This work constitutes advances in (1) real-time observing in aquatic ecosystems and (2) sensor actuated sampling for biological analysis.

Approach

The goal of developing a predictive understanding of aquatic microorganismal distributions warrants a continuous (sensing) presence in the environment to enable real-time acquisition and analysis of chemical and physical data collected at relevant spatiotemporal scales, and correlated with measurements of specific microorganisms. However, at the scales required to attain this goal, it is infeasible to deploy a set of stationary monitoring stations that will provide sufficient spatial density and continuous monitoring. Conversely, deploying a fleet of mobile autonomous vehicles might provide adequate spatial coverage but insufficient temporal coverage. The concept of deploying a high-density, wireless network consisting of both stationary and mobile components to aid each other has been recently introduced. Stationary buoys provide low-resolution spatial sampling with high temporal resolution while a mobile robotic boat provides high-resolution spatial sampling with relatively low temporal resolution. Collectively, we believe this network provides unprecedented coverage and thus unique insights into microbial plankton distributions and dynamics. Here we describe our prototype sensor-actuator network consisting of 10 buoys and a robotic boat, equipped with a collection of simple, off-the-shelf sensors (GPS, thermistors, fluorometers) that can be deployed in-situ to gather and analyze relevant data in an aquatic environment. We describe the design of the system. Data collected from preliminary field trials are described in the section on aquatic applications.

Systems/Experiments

The stationary nodes (buoys) continuously monitor the aquatic environment at the location at which they are deployed, and communicate the collected sensor information to the robotic boat, which is capable of autonomous navigation and sampling. The robotic boat is a modified RC airboat. An air propeller provides propulsion and minimizes disturbance to the water. For a description of the hardware constituents of the system see the project website – we omit it here for brevity. Our software system is built atop EmStar, a software system for developing and deploying wireless sensor networks involving Linux-based platforms. There are three principal components – a data logger, an interface between the network and the users, and a toolkit to visualize data. The boat is directed by information collected and processed within the network to identify features of biological interest. The stargate board on the boat receives and processes the inputs from GPS, compass, sensors and the network, makes decisions, and sends appropriate navigation commands to the navigation module. The basic stamp in the navigation module converts these into appropriate commands for the motor controllers, which are connected to the rudder and the propeller. By sending appropriate commands, the boat can navigate in both forward and reverse directions. The robotic boat operates in three modes. In the radio-controlled mode, the boat is controlled using RF control from the shore. In the computer-controlled mode, appropriate control commands can be sent from the base station to the boat over the ad-hoc network. In the autonomous mode, the boat uses GPS waypoint locations (set by a human user or the buoy network) and sensor information to compute control commands.

Many environmental applications require spatiotemporal sampling in the 3D extent of the complete environment. This requires proper distribution of sensing assets at the site of measurement. Since sensing resources are relatively sparse in distribution, proper deployment design is critical. However, such design is constrained by the limitations on time and resources. Further, it is only at the time of deployment that the required spatial distribution of sensors is known (further, this optimal distribution may itself be time-varying). Thus, a new method is under development – multiscale iterative design – that takes advantage of the full set of CENS actuated sensing resources supported by statistical computing methods. Iterative design is guided by a hierarchy of sensor data sources, beginning with static sources and the augmented by actuated sensing, itself guided by iterative design.

As an example of an application of current interest for CENS Aquatic research, this includes the analysis the microbial organism concentration in fresh water systems. Here, the application objective is the reconstruction of 3D models of temperature, contaminant nutrients, and chlorophyll distribution, determining both the environment supporting growth and the concentration of biomass itself.

Figure 1: The NAMOS/NIMS system experimental architecture planned for deployments in lake systems. The currently deployed NAMOS buoys are shown at left at Lake Fulmor. The results of multiscale iterative design will determine the rapid deployment of NIMS actuated systems at right.

A current example experimental system, shown in Figure 1, is directed to sampling of the Lake Fulmor system at the James Reserve for reconstruction of temperature and chlorophyll concentration distribution. The enabling ENS platforms include static sensor buoys that are augmented by two forms of 2D actuated sensor systems. Buoy nodes are deployed across the surface of the lake and provide a spot measurement of temperature and chlorophyll. This data can be used for an estimation of these parameters in the complete extent of the lake. 2D systems, including deployed NIMS devices, are guided by the analysis of buoy data that provides not only a reconstruction of field variables, but, also a directive that determines the location of NIMS deployment providing the greatest enhancement in fidelity of reconstruction. Once deployed, NIMS provides highly accurate sampling in the plane orthogonal to the surface plane in which the buoys are deployed. The combined results from these systems will provide a 3D reconstruction of field variables using an approach that minimizes deployment time and ensures the lowest latency response to sampling demands.

Accomplishments

Autonomous navigation over water is non-trivial since wind and water currents affect the boat’s heading and speed. Limited GPS availability and inaccuracies in sensor information (both GPS and compass) introduce further problem and are an area of ongoing research. We use a PID controller to compensate for disturbances and sensor errors while performing waypoint following. Based on the accuracy of the GPS, the system dynamically adjusts its error tolerances for waypoints resulting in reliable waypoint following in varying conditions. Figure 2 gives a high-level pseudocode description of the way-point navigation and control algorithms and Figure 3 shows a trace of the boat executing a radiator pattern in water under the control of these algorithms.

Fig. 2: Algorithm 1 is the main loop for navigation and sampling which makes calls to Algorithms 2 (location tracking), 3 (heading tracking), and 4 (rudder control).

Fig. 3: A radiator pattern navigation trace for the boat operating in autonomous waypoint-following mode.

Future Directions

Our most important future direction to perform several measurement campaigns in Lake Fulmor and other freshwater sources. As described elsewhere in this report we are also actively pursuing a combined NIMS and NAMOS measurement campaign. A major area of focus will also be the coordination of multiple robotic boats for sampling.

People

• Students: Amit Dhariwal, Bin Zhang, Arvind Pereira
• Staff: Carl Oberg and Beth Stauffer
• Faculty: David Caron and Gaurav S. Sukhatme