Publications

We present a bio-inspired control framework that enables autonomous upstream swimming in robotic fish through multisensory feedback integration. By combining flow sensing, proprioception, and visual odometry, our system achieves self-organized rheotaxis behavior without explicit flow models.

This work uncovers the mechanism of vortex-induced propulsion in upstream swimming, demonstrating how robots can exploit environmental flow structures for energy-efficient locomotion. We show that physical intelligence emerges from morphological computation without explicit control strategies.

We introduce Aquarium 2.0, a differentiable simulation framework that enables gradient-based optimization of aquatic robot designs. The simulator combines high-fidelity fluid dynamics with soft body mechanics, enabling co-design of morphology and control.

We present a hybrid aerial-aquatic robot capable of seamless transition between flight and swimming. The key innovation is a variable stiffness propulsion module that adapts to different fluid environments, enabling efficient locomotion in both air and water.

We develop a reduced-order Cosserat rod model that achieves real-time simulation of soft continuum robots. By exploiting geometric structure and modal decomposition, our method enables interactive control and planning for soft manipulators.

We develop a vision-based tactile sensor for measuring fluid flow patterns on robot surfaces. Inspired by GelSight technology, our sensor provides high-resolution flow velocity fields that enable reactive swimming control.

This paper presents a novel control strategy for soft aquatic robots that modulates material stiffness both spatially and temporally. We demonstrate significant improvements in swimming efficiency through synchronized stiffness variation with undulatory motion.

We survey and categorize bio-inspired variable stiffness mechanisms across multiple biological systems, providing design principles for soft robotic applications. The work includes experimental characterization of several novel stiffness modulation approaches.

This work analyzes the hydrodynamics of multi-joint undulatory propulsion using both computational fluid dynamics and experimental validation. We identify optimal joint configurations for maximizing thrust generation in BCF swimming.

Award-winning senior design project developing an autonomous surface vehicle with bio-inspired variable stiffness propulsion. The system integrates perception, planning, and adaptive control for efficient marine navigation.

We develop a computational framework for simulating flexible appendages in fluid environments using discrete variational integrators. The method preserves geometric structure and energy conservation properties, enabling long-time stable simulations.

This paper introduces a novel approach to modeling body-caudal fin (BCF) swimming using structure-preserving numerical methods. We demonstrate improved accuracy and computational efficiency compared to traditional finite element approaches.

We present a salamander-inspired robot capable of seamless transition between terrestrial and aquatic locomotion. Central pattern generators (CPGs) enable adaptive gait modulation based on environmental feedback.

We develop a pneumatically-actuated soft gripper with variable stiffness capabilities for underwater manipulation. The gripper adapts its compliance to handle fragile marine organisms and rigid objects with equal dexterity.

This work implements the C-start escape response observed in fish using a biomimetic robotic platform. We demonstrate high-acceleration maneuvers achieving performance comparable to biological systems.