ARchaeological RObot systems for the World's Seas (ARROWS) EU Project proposes to adapt and develop low-cost Autonomous Underwater Vehicle (AUV) technologies to significantly reduce the cost of archaeological operations, covering the full extent of archaeological campaign. ARROWS methodology is to identify the archaeologists requirements in all phases of the campaign and to propose related technological solutions. Starting from the necessities identified by archaeological project partners in collaboration with the Archaeology Advisory Group, a board composed of European archaeologists from outside ARROWS, the aim is the development of a heterogeneous team of cooperating AUVs capable of comply with a complete archaeological autonomous mission. Three new different AUVs have been designed in the framework of the project according to the archaeologists' indications: MARTA, characterized by a strong hardware modularity for ease of payload and propulsion systems configuration change; U-CAT, a turtle inspired bio-mimetic robot devoted to shipwreck penetration and A Size AUV, a vehicle of small dimensions and weight easily deployable even by a single person. These three vehicles will cooperate within the project with AUVs already owned by ARROWS partners exploiting a distributed high-level control software based on the World Model Service (WMS), a storage system for the environment knowledge, updated in real-time through online payload data process, in the form of an ontology. The project includes also the development of a cleaning tool for well-known artifacts maintenance operations. The paper presents the current stage of the project that will lead to overall system final demonstrations, during Summer 2015, in two different scenarios, Sicily (Italy) and Baltic Sea (Estonia).

Motion planning is one of the main problems studied in the field of robotics. However, it is still challenging for the state-of-the-art methods to handle multiple conditions that allow better paths to be found. For example, considering joint limits, path smoothness and a mixture of Cartesian and joint-space constraints at the same time, pose a significant challenge for many of them. This work proposes to use Timed-Elastic Bands for representing the manipulation motion planning problem, allowing to apply continuously optimized constraints to the problem during the search for a solution. Due to the nature of our method, it is highly extensible with new constraints or optimization objectives. The proposed approach is compared against state-of-the-art methods in various manipulation scenarios. Results show that it is more consistent and less variant while performing in a comparable manner to the state-of-the-art. This behaviour allows the proposed method to set a lower bound performance guarantee for other methods to build upon.

No abstract

No abstract

Learning from Demonstration (LfD) is a family of methods used to teach robots specific tasks. It is used to assist them with the increasing difficulty of performing manipulation tasks in a scalable manner. The state-of-the-art in collaborative robots allows for simple LfD approaches that can handle limited parameter changes of a task. These methods however typically approach the problem from a control perspective and therefore are tied to specific robot platforms. In contrast, this paper proposes a novel motion planning approach that combines the benefits of LfD approaches with generic motion planning that can provide robustness to the planning process as well as scaling task learning both in number of tasks and number of robot platforms. Specifically, it introduces Dynamical Movement Primitives (DMPs) based LfD as initial trajectories for the Stochastic Optimization for Motion Planning (STOMP) framework. This allows for successful task execution even when the task parameters and the environment change. Moreover, the proposed approach allows for skill transfer between robots. In this case a task is demonstrated to one robot via kinesthetic teaching and can be successfully executed by a different robot. The proposed approach, coined Guided Stochastic Optimization for Motion Planning (GSTOMP) is evaluated extensively using two different manipulator systems in simulation and in real conditions. Results show that GSTOMP improves task success compared to simple LfD approaches employed by the state-of-the-art collaborative robots. Moreover, it is shown that transferring skills is feasible and with good performance. Finally, the proposed approach is compared against a plethora of state-of-the-art motion planners. The results show that the motion planning performance is comparable or better than the state-of-the-art.