Co Manipulation
Co-manipulation research focuses on enabling robots to collaboratively manipulate objects with humans, aiming for safe and efficient interaction. Current efforts concentrate on developing robust methods for estimating human intent in real-time, often employing dynamic systems models and particle filters, and decomposing interaction forces to better understand and control collaborative actions. This involves utilizing techniques like Dynamic Movement Primitives (DMPs) and Convolutional Neural Networks (CNNs) for motion planning and object state estimation, leading to improved control strategies for both rigid and soft materials. The ultimate goal is to create more intuitive and effective human-robot collaboration in various applications, from industrial automation to assistive robotics.