Artistic pattern generation by multiple robots

The research on pattern generation problems has received a considerable attention in recent decades. The problems arise in the security, surveillance, search and rescue in disaster areas. In these applications, it is required to generate given pattern autonomously by multiple robots. We tackle this problem based on the centroidal Voronoi tessellation (CVT), by first planning a configuration pf robot which mimics the given pattern and driving each robot to the desired position to realize the planned configuration.

 

 

Adaptive allocation for low earth orbit satellite constellations

This work proposes a strategy of adaptive and efficient coverage control for allocating low earth orbit satellites systems with the function to avoid obstacle which refers to the existence of time-varying uncertainties. An optimized allocation approach is accomplished by an optimal partition strategy of Spherical Centroidal Voronoi Tessellation (SCVT). For the control method to realize the SCVT in outer space environment, an adaptive motion controller is constructed. The simulation has tested that the controller has legitimate stability and its rationality in a spherical surface scenario is solid.

 

 

Vision-based flood monitoring by a swarm of robots

This research is concerned with developing control strategies for tracking the propagation of a flood area by using a group of unmanned aerial vehicles (UAVs). UAVs are required to detect the inundation regions and allocate among the interior region of the flood area, covering the region as much as possible with less overlapping of the UAVs’ field of vision. Corresponding control algorithms will be proposed for the aforementioned types of UAVs to implement the control strategy. The feasibility of the control strategy is verified under simulations.

 

 

Safe autonomous driving

In recent years, there has been a growing interest in the research of autonomous driving. It allows vehicles to operate automatically based on artificial intelligence, computer vision, etc. The introduction of autonomous driving enables traffic conditions to be better monitored in large scale, thereby reducing the possibility of traffic jams. Besides, autonomous driving can liberate people from steering wheels, making them spend their commuting time more efficiently. However, Safety (collision avoidance), an essential issue in autonomous driving, remains unsolved. We aim to provide collision-free algorithm in chaotic traffic condition for multiple vehicles.