Publications
Clicking on any of the links below will redirect you to the abstract and details of my contributions.
Published in International Conference on Robotics and Automation in Industry (ICRAI), 2021
This paper proposes a decentralized formation control approach for the swarm of Unmanned Aerial Vehicles (UAVs). The system requires no predefined leaders and followers, and only local information is needed to control the individual UAV. In the proposed algorithm, each UAV motion constitutes local information such as pose and velocities shared with several other closest neighbors using the Nearest Neighbour Search (NNS) algorithm. The proposed approach leveraged a 3D mapping technique known as OctoMap for mapping unknown terrain and uses a variant of Rapid Exploration Random Trees (RRT*) algorithm with path length minimization as optimization objective for UAVs swarm path planning and navigation. We found that our proposed approach can be applied in a hazardous environment where an extensive area needs to be covered for search and rescue missions. The proposed algorithm is tested and validated with three different experiments on a dynamic simulator showing promising results.
Published in 2017 IEEE World Haptics Conference (WHC), Germany, 2017
While teleoperation provides a possibility for a robot to operate at extreme conditions instead of a human, teleoperating a robot still demands a heavy mental workload from a human operator. Learning from demonstrations can reduce the human operator’s burden by learning repetitive teleoperation tasks. However, one of challenging issues is that demonstrations via teleoperation are less consistent compared to other modalities of human demonstrations. In order to solve this problem, we propose a learning scheme based on Dynamic Movement Primitives (DMPs) which can handle less consistent, asynchronized and incomplete demonstrations. In particular we proposed a new Expectation Maximization (EM) algorithm which can synchronize and encode demonstrations with temporal and spatial variances, different initial and final conditions and partial executions. The proposed algorithm is tested and validated with three different experiments of a pegin-hole task conducted on 3-Degree of freedom (DOF) masterslave teleoperation system.
