This work demonstrates the feasibility of near real-time, plenoptic-inertial navigation on a low-cost central processing unit (CPU). To enable real-time operation, a standard plenoptic camera was modeled as a system of stereo cameras and triangulation was used to estimate relative pose from a minimal set of sub-aperture images. The relationship between distance and disparity for the simplified model was experimentally validated in an aquatic environment, using a first-generation Lytro camera, and a mean error of 2% of the target distance was obtained. This paper culminates with testing the proposed navigation system on an in-house developed, novel, biologically inspired, autonomous underwater vehicle (AUV), CephaloBot. The test consisted of the AUV rotating around a static object while maintaining a fixed separation distance. The mean position error from the test was 2.5% of the target distance. With the simplified plenoptic model, only 750 ms were required to process the raw plenoptic data and estimate position on an Intel i5 CPU. The processing delay was short enough that the delayed position measurements bounded the effects of sensor drift when fused with an inertial measurement unit using a delayed extended Kalman filter. This result demonstrates the feasibility of plenoptic-inertial navigation on a low-cost CPU.

This article presents a variable buoyancy system that uses cephalopod-inspired, fiber-reinforced elastomer membranes as bladders to enable model-based depth and pitch control of an autonomous underwater vehicle. The fiber reinforcement allows for a single measurement of each membrane to fully define the geometry of the membrane, regardless of the membrane’s orientation or level of inflation. This allows for precise control of each bladder’s enclosed water mass and center of mass by adjusting the differential pressure between the inside of the bladder and the vehicle’s cabin using water from the environment as the working fluid. Positioning bladders at the fore and aft of the vehicle enables control of both the vehicle’s density and center of mass while minimizing center of mass disturbances due to sloshing in and gravitational loads on the membranes. A nonlinear, adaptive, backstepping, trajectory-tracking controller is presented and proven to be asymptotically stable with Lyapunov stability analysis. The advantages of the proposed variable buoyancy system and controller are demonstrated through simulation and validated with 2 degree of freedom depth and pitch step and trajectory-tracking experiments in a large water tank. We evaluate the performance of the bladders and controller on the CephaloBot, an autonomous underwater vehicle designed and built by our group.

In this letter, we present a distributed pressure sensory system, inspired by the lateral line found in fish, for estimating hydrodynamic forces acting on an autonomous underwater vehicle. By canceling these forces using the vehicle control system, the dynamics of the vehicle can be decoupled from the state of the surrounding fluid. We discuss a control scheme which combines the measured hydrodynamic forces with a feedback controller that is robust to disturbances and measurement errors, achieving asymptotic tracking in the presence of bounded disturbances and errors. We compare this robust lateral line controller to a baseline robust controller not using the lateral line system and to a controller using the lateral line with simple PD feedback. We simulate a spatially and time varying velocity flow field on the scale of the vehicle and show that the robust lateral line controller has improved performance compared to both the baseline robust controller and the PD controller using the lateral line system.