Shape Servoing

Shape servoing aims to control the shape of deformable objects using robotic manipulation, a challenging task due to the objects' complex, non-rigid dynamics. Current research focuses on developing robust control algorithms, such as sliding mode control and model predictive control, often incorporating neural networks (e.g., for goal learning or occlusion compensation) to improve performance and handle uncertainties in object properties and sensor data. These advancements are crucial for automating tasks involving deformable objects in various fields, including surgery and manufacturing, where precise shape control is essential. The development of model-free approaches and efficient data utilization techniques are also key areas of investigation.