From the Netherlands to Italy to advance robotic manipulation in complex environments

In January and February 2026, Yaolei Shen, PhD candidate at the University of Twente, completed his second secondment at the Department of Information Engineering at the University of Pisa, in collaboration with the “E. Piaggio” Research Center, as part of the NeutraWeed project.

Building on his previous secondment in September-October 2025, which focused on the development of motion planning algorithms for deformable object manipulation, this new research period was dedicated to the experimental validation of the proposed trajectory optimization framework.

Yaolei carried out extensive experimental activities to test the algorithm in more advanced and realistic scenarios. With the support of the host research team, he contributed to the development of multiple experimental setups, including systems where two robotic arms collaboratively manipulate deformable objects. In addition, a novel tethered aerial-terrestrial robotic platform was established, enabling the evaluation of the algorithm in cluttered environments with obstacle avoidance.

This platform is intended for applications such as weed detection. Compared with conventional robotic systems, the aerial component provides a wider field of view and enhanced mobility and accessibility in unstructured outdoor environments, while the terrestrial component supplies power and computational support. Furthermore, the developed global motion planning algorithm ensures system safety when operating in dense vegetation.

The results confirm and extend the findings of the previous secondment: the developed algorithm is capable of generating dynamically feasible and collision-free trajectories for the manipulation of continuous deformable objects across different robotic platforms and complex environments. The activities also produced a range of materials, including algorithm code and experimental datasets, which will be made publicly available.

Overall, this secondment represents a natural continuation of the earlier work, reinforcing collaboration between the University of Twente and the University of Pisa, and advancing research on robotic manipulation in complex, real-world scenarios.