Search

Estimating the pose of multiple animals is a challenging computer vision problem: frequent interactions cause occlusions and complicate the association of detected keypoints to the correct individuals, as well as having highly similar looking animals that interact more closely than in typical multi-human scenarios. To take up this challenge, we build on DeepLabCut, an open-source pose estimation toolbox, and provide high-performance animal assembly and tracking-features required for multi-animal scenarios. Furthermore, we integrate the ability to predict an animal's identity to assist tracking (in case of occlusions). We illustrate the power of this framework with four datasets varying in complexity, which we release to serve as a benchmark for future algorithm development., DeepLabCut is extended to enable multi-animal pose estimation, animal identification and tracking, thereby enabling the analysis of social behaviors.

In fish, the tail is a key element of propulsive anatomy that contributes to thrust during swimming. Fish possess the ability to alter tail stiffness, surface area and conformation. Specifically, the region at the base of the tail, the caudal peduncle, is proposed to be a key location of fish stiffness modulation during locomotion. Most previous analyses have focused on the overall body or tail stiffness, and not on the effects of changing stiffness specifically at the base of the tail in fish and robotic models. We used both computational fluid dynamics analysis and experimental measurements of propulsive forces in physical models with different peduncle stiffnesses to analyze the effect of altering stiffness on the tail angle of attack and propulsive force and efficiency. By changing the motion program input to the tail, we were able to alter the phase relationship between the front and back tail sections between 0° and 330°. Computational simulations showed that power consumption was nearly minimized and thrust production was nearly maximized at the kinematic pattern where φ = 270°, the approximate phase lag observed in the experimental foils and in free swimming tuna. We observed reduced thrust and efficiency at high angles of attack, suggesting that the tail driven during these motion programs experiences stalling and loss of lift. However, there is no single peduncle stiffness that consistently maximizes performance, particularly in physical models. This result highlights the fact that the optimal caudal peduncle stiffness is highly context dependent. Therefore, incorporating the ability to control peduncle stiffness in future robotic models of fish propulsion promises to increase the ability of robots to approach the performance of fish.

We describe the implementation of a versatile large-eddy simulation (LES) model into an immersed boundary solver and its application to study of swimming and flying in nature. Specifically, we employ the dynamic global coecient 1 form of the algebraic eddy viscosity model introduced by Vreman. 2 This model is implemented within a Cartesian grid based IBM amenable to flows with moving boundaries. The implementation is then validated against canonical flows including turbulent channel flow and flow past a circular cylinder. The validated methodology is then applied to the pectoral fin driven yaw turn maneuver of a bluegill sunfish. This flow is typical of biological flows in that the geometry involved are highly complex, three-dimensional and non-stationary and the flow contains laminar, transitional and turbulent zones. The results obtained clearly demonstrate the versatility of the present LES approach for studying complex flows associated with flying and swimming in nature.

his work is part of a research program designed to develop a maneuvering propulsor for an Autonomous Underwater Vehicle (AUV’s) based on the mechanical design and performance of fish fin. Fish are notable for their ability to maneuver and to accurately position themselves even in highly unsteady flows. Fish can maneuver in tight spaces and accelerate and decelerate quickly from rest or low speed with the same propulsion system. Also, fish fins are remarkable fluid control devices with supporting bony elements that are under active muscular control. In addition to muscular actuation at the base of the fin which governs overall fin position, fish fins possess a bilaminar design that allows active control of propulsor surface curvature (Lauder et al., 2005). This paper will focus on hydrodynamic analyses of bluegill sunfish’s pectoral fin which were carried out to guide the design of a flexible propulsor for AUV’s.

A computational study of fish fin hydrodynamics is being conducted concurrently with experimental analysis. The presence of a complex shaped moving boundary makes this a difficult proposition for computational fluid dynamics and here we use a Cartesian grid based immersed boundary solver designed to handle such flows in all their complexity. High resolution, high speed video of fish fin movement during steady swimming by a bluegill sunfish (Lepomis macrochirus) is obtained and these are used as a basis for developing a high fidelity geometrical model of the flapping fish fin. Simulations are carried out in order to examine the hydrodynamic performance of the fish fin and understand the wake topology of such fins. The fin motion is highly complex and a number of different strategies including proper orthogonal decomposition of the fin kinematics are used to examine the various kinematical features of the fin motion as well as their impact on the fin performance.