Chunhui Li
Bilateral Comparison Confirms NIM’s and NIST’s Gas Flow Capabilities

A bilateral comparison of the gas flow standards of the National Institute of Metrology (NIM) in China and the National Institute of Standards and Technology (NIST) in the USA was conducted from June 2008 to October 2009. Two critical flow venturis (CFVs) with nominal throat diameters of 10 mm and 20 mm, respectively, were selected as transfer standards. The CFVs were calibrated on NIM’s 20 m³ PVTt system and then on NIST’s 26 m³ PVTt system. The results demonstrate the equivalence between NIM’s and NIST’s gas flow measurement capabilities for flows ranging from 1000 liters per minute to 4000 liters per minute at reference conditions of 101.325 kPa and 293.15 K. The experimental data agreed with existing theoretical models within 0.07 %. All the data agree with the ISO 9300 empirical equation within its 0.3 % expanded uncertainty limit.

Matti Saarikko, Jani Koskimaa, Pasi Penttinen, Jussi Ali-kovero, Jürgen Köberle, Stefan Bullmer, Jens Henzler, Björn Steurer, Dr. Alexander Wolf, Steffen Ruedel, Fabian Meilinger, Christian Seidl, Matthias Hentschel
ELROB 2018 - Team Patria participating in Convoy Scenario

The team Patria consists of three project partners with different focus in research and development. The Patria AMV vehicles used in the scenario are both R&D platforms, which have been developed in a cooperation of Patria and Diehl Defence. The platforms are used by the team in the convoy competition at the ELROB 2018. While one of the vehicles (called AMV 2) is equipped with a Drive-by-Wire system developed by Patria and with a sensor package to track a leading vehicle the other AMV (AMV 1) is a vehicle with standard configuration. Only a LTB (Leader Tracking Box) developed by Diehl Defence is installed on this vehicle. The LTB consists of a GPS/IMU system and radio. Therefore, the LTB is able to determine its position and to send it to the following AMV 2 vehicle. The AMV 2 additionally has a Continental automotive RADAR which is used for tracking. Both, sensor data from the RADAR, and the LTB position estimation, are fused to achieve a robust and accurate tracking of the leading vehicle.

V. Gradetsky, M. Knyazkov
Unmanned vehicle systems in unstructured environments: challenge and current status

The scientific and applied interest to Unmanned vehicle systems (UVS) or Unmanned guided vehicles (UGV) is growing every year and a lot of world-known universities and companies are working very hard in R&D of UVS. The fully autonomous vehicles or ones with different degrees of autonomy vehicles are used on the land, in the air, and in the sea for broad range of technologies including hazardous waste cleanup, agriculture crops processing, transportation, underwater monitoring, nuclear power station repair, as well as security, inspection, demining, military operations.The new uses for these systems may be found every day.
Real UVS has to understand control commands fast and without mistakes. They have to possess logically thinking, have possibilities of learning and training, produce decision making by themselves, to move in unknown environment without operator. However, today in practice the situation may arise when operator forced to intervene in teleoperated control because of unsufficient reliability of robot behaviour. High level of artificial intelligence and mind as well as data base and special software are required for real UVS.

F. Ebert, P. Berthold, P. Burger, T. Engler, A. Frericks, B. C. Heinrich, J. Kallwies, M. Kusenbach, T. Luettel, K. Metzger, M. Michaelis, B. Naujoks, A. Sticht, H.-J. Wuensche
ELROB 2018 – Convoy and Mule of Team MuCAR

We present the hard- and software components of team MuCAR’s fully autonomous vehicles to participate at the ELROB 2018 in the convoy and mule scenarios. For the convoy scenario, different tracking approaches are applied to track the leading vehicle. Data association of the tracking results is done in a PHD filter framework. Given the resulting estimate, an optimization-based planning module computes kinematically feasible trajectories to follow the leading vehicle’s path as close as possible with a velocity-dependent lateral distance.
In the mule scenario, the leading guide is tracked either with a LiDAR-based Greedy Dirichlet Process Filter (GDPF) approach or in a vision-only approach by segmenting the disparity image and reprojection into 3D space to match the existing track. During the shuttling phase, two environment modeling algorithms were implemented. Again, one mapping approach is based on LiDAR and the second is based on vision only. The LiDAR mapping approach includes besides occupancy, color information, heights and terrain slopes. In the vision-only mapping approach a dense disparity image with a tri-focal camera is generated and back-projected to create a virtual 3D scene.
Finally, a high-level mission planning module and a local trajectory planner are used for GPS-based autonomous shuttling. The local trajectory planner based on a hybrid A* approach incorporates data from the environment mapping modules for goal-oriented navigation and local obstacle avoidance.

Niels Nauwynck, Haris Balta, Geert De Cubber, Hichem Sahli
In-flight launch of unmanned aerial vehicles

This paper considers the development of a system to enable the in-flight-launch of one aerial system by another. The paper will discuss how an optimal release mechanism was developed, taking into account the aerodynamics of one specific mother and child UAV. Furthermore, it will discuss the PID-based control concept that was introduced in order to autonomously stabilize the child UAV after being released from the mothership UAV. Finally, the paper will show how the concept of a mothership UAV + child UAV combination could be usefully taken into advantage in the context of a search and rescue operation.