Development of numerical models of a bio-robotic sea lion to Investigate the robot's swimming and maneuvering strategies

Shraman Kadapa

doi:10.17918/00010994

There has been a significant rise in the development of underwater robots that emulate the locomotion strategies of biological systems. As these robots gain additional degrees of freedom and become increasingly complex, it gets difficult to understand how the actuation of various body segments and how their coordinated motion contributes to swimming and maneuvering performance. To address these challenges, numerical modeling is essential for predicting performance and guiding the design and development of such robots. This thesis is divided into two main phases. The first phase focuses on the development of complementary numerical models of a bio-robotic sea lion to investigate the robot's swimming and maneuvering strategies. Two distinct models were developed: (1) In the Euler-Poincare model, the equations of motion were derived in closed-form that provided a full mathematical representation of the robot's dynamics. This model supported stability analysis and other energy based evaluations. (2) A Newton-Euler based formulation was developed in Simscape which enabled fast numerical simulations and good visualization of the robot's kinematics. This model was used extensively to study maneuverability and served as a training environment for reinforcement learning-based gait development. Hydrodynamic coefficients in both models were estimated using a combination of computational fluid dynamics (CFD), strip theory, and refined using a genetic algorithm to align the coefficients with the bio-robotic sea lion's experimental behavior. Using this framework, simulations were conducted to analyze how the angular positioning of the head, pelvis, foreflippers, and hind flippers influence turning behavior. The second phase of the thesis investigated how the coordination between two flapping fins affected propulsion. A two-dimensional CFD simulation was developed to study how geometric (spacing) and temporal (phase and frequency) relationships between the fins influence the generated thrust and flow structures. Validation of the numerical model against experimental data showed strong agreement when center of mass trajectories during swimming and maneuvering trials were compared. Results from the maneuverability study revealed that simultaneous actuation of the head, pelvis, and foreflippers yielded the most effective turning performance, achieving higher angular velocities, and tighter turning radii. The pelvic section and hind flippers, located posterior to the robot's center of mass, contributed more significantly to turns than the head, although the head did aid in turns. In the fin interaction study, it was found that the spacing, flapping frequency, and phase difference between fins are highly coupled and critically influence thrust production. Fins placed too close to each other experienced disrupted flow and reduced thrust, while fins spaced one body length apart and flapped with a smooth upstream-to-downstream wake transition produced the strongest propulsion. Collectively, the findings from both studies underscore the importance of coordinated actuation of different body segments to produce propulsive forces and enhanced maneuverability. These insights provide a foundation for advancing the design, development and control of next generation bio-inspired underwater robotic systems.

Development of numerical models of a bio-robotic sea lion to Investigate the robot's swimming and maneuvering strategies

Files and links (1)

Abstract

Metrics

Details

Development of numerical models of a bio-robotic sea lion to Investigate the robot's swimming and maneuvering strategies

Files and links (1)

Abstract

Metrics

Details

Drexel University Social media