Research
Current Projects
Fin Actuated Autonomous Underwater Vehicles >
Inspired by nature, our intent is to generate novel bio-inspired systems that can out-perform existing engineered systems in speed, agility and efficiency. We focus on bioinspired actuators (based on fish-fin type structures) to control fluid dynamic artifacts (both in and away from the boundary layer) that will ultimately affect speed, agility, and stealth of air and underwater autonomous vehicles. Our current work focuses on refining the model and extending it to more aggressive operation such as for flexible foils. Photo and Video Gallery >
Coordination and Cooperation in Multivehicle Systems >
By modeling planar vehicle motion using Frenet-Serret formulas and 3D motion using natural Frenet frames, nonholonomic constraints and actuation bounds (forward velocity and turning rate) can be incorporated into vehicle dynamics.
Hierarchical Integrated Communication and Control >
The objective of the project is to design integrated control and communication algorithms that guarantee that a set of vehicles with differing data capabilities will conform to a specified spatial distribution. Each vehicle is equipped with physical devices that provide local information from sensing and global information from communication. Sensing devices are assumed to be lower cost in terms of power, computation, and range of operation, while communication devices are comparatively higher cost but with more information density and reconfigurability.
Coordination, Steering and Deconfliction for Unmanned Air Vehicles >
The focus of the work in this project is to develop coordinated control algorithms for formations of UAVs (Unmanned Air Vehicles) that allow the formation to track single or multiple targets while preventing collisions. Successful tracking of a single target will be characterized by the centroid of the group tracking the position of the target, and tracking of multiple targets will be characterized by both the centroid of the group tracking the centroid of the set of targets and the footprint of the group covering the set of targets.
Modeling and Control in Mixed Human/Robotic Teams >
This project is a five-year multi-university research effort aimed at understanding key aspects of cooperative distributed decision making, coordination, and distributed control of groups of humans and autonomous machines.
Integrated Control, Networking and Digital Communication
The focus of this project is design and analysis of integrated communication networks in service of coordinated control of multi-vehicle systems. To achieve globally desirable formation behavior, the controller of each vehicle must respond to the motion and state of others. Each vehicle is equipped with wireless radio to disseminate vehicle state information. The intellectual merit of the project consists of three interdependent theoretical components led respectively by the co-PIs: (1) nonlinear coordinated control over dynamic graphs, (2) cross-layer optimization of wireless networks, and (3) physical layer solutions to decentralized communication and control. An overarching experimental component will be used to compare and test how the communication solutions provided by (2) and (3) perform under identical primitives from (1). This system will first use the existing 3D indoor autonomous underwater vehicle testbed at the University of Washington and later the in progress open water acoustic Seaglider system.
Past Projects
Schooling in Nature and Engineering
The work in this proposal focuses on a novel development of systems-level biomimetic control-theoretic models and motion control algorithms for embedded, emergent coordinated behavior of multiple organisms. To obtain data for traffic rules used by fish operating under heterogeneous sensing and traffic rules, schooling experiments are being performed using giant danio with different internal states to impose varying sensing rules. Fish trajectories are tracked with cameras, and the data is then extracted for use in derived control theoretic algorithms to be applied to engineered systems. These algorithms will be implemented on a testbed of free swimming remote-controlled robots. Experiments on the robot testbed will be used to validate and refine models and motivate further testing criteria for the schools of fish.