@inproceedings{NikolovskiRekeElsenetal.2021, author = {Nikolovski, Gjorgji and Reke, Michael and Elsen, Ingo and Schiffer, Stefan}, title = {Machine learning based 3D object detection for navigation in unstructured environments}, series = {2021 IEEE Intelligent Vehicles Symposium Workshops (IV Workshops)}, booktitle = {2021 IEEE Intelligent Vehicles Symposium Workshops (IV Workshops)}, publisher = {IEEE}, isbn = {978-1-6654-7921-9}, doi = {10.1109/IVWorkshops54471.2021.9669218}, pages = {236 -- 242}, year = {2021}, abstract = {In this paper we investigate the use of deep neural networks for 3D object detection in uncommon, unstructured environments such as in an open-pit mine. While neural nets are frequently used for object detection in regular autonomous driving applications, more unusual driving scenarios aside street traffic pose additional challenges. For one, the collection of appropriate data sets to train the networks is an issue. For another, testing the performance of trained networks often requires tailored integration with the particular domain as well. While there exist different solutions for these problems in regular autonomous driving, there are only very few approaches that work for special domains just as well. We address both the challenges above in this work. First, we discuss two possible ways of acquiring data for training and evaluation. That is, we evaluate a semi-automated annotation of recorded LIDAR data and we examine synthetic data generation. Using these datasets we train and test different deep neural network for the task of object detection. Second, we propose a possible integration of a ROS2 detector module for an autonomous driving platform. Finally, we present the performance of three state-of-the-art deep neural networks in the domain of 3D object detection on a synthetic dataset and a smaller one containing a characteristic object from an open-pit mine.}, language = {en} } @article{SchulteTiggesFoersterNikolovskietal.2022, author = {Schulte-Tigges, Joschua and F{\"o}rster, Marco and Nikolovski, Gjorgji and Reke, Michael and Ferrein, Alexander and Kaszner, Daniel and Matheis, Dominik and Walter, Thomas}, title = {Benchmarking of various LiDAR sensors for use in self-driving vehicles in real-world environments}, series = {Sensors}, volume = {22}, journal = {Sensors}, number = {19}, publisher = {MDPI}, address = {Basel}, issn = {1424-8220}, doi = {10.3390/s22197146}, pages = {20 Seiten}, year = {2022}, abstract = {Abstract In this paper, we report on our benchmark results of the LiDAR sensors Livox Horizon, Robosense M1, Blickfeld Cube, Blickfeld Cube Range, Velodyne Velarray H800, and Innoviz Pro. The idea was to test the sensors in different typical scenarios that were defined with real-world use cases in mind, in order to find a sensor that meet the requirements of self-driving vehicles. For this, we defined static and dynamic benchmark scenarios. In the static scenarios, both LiDAR and the detection target do not move during the measurement. In dynamic scenarios, the LiDAR sensor was mounted on the vehicle which was driving toward the detection target. We tested all mentioned LiDAR sensors in both scenarios, show the results regarding the detection accuracy of the targets, and discuss their usefulness for deployment in self-driving cars.}, language = {en} } @inproceedings{NikolovskiLimpertNessauetal.2023, author = {Nikolovski, Gjorgji and Limpert, Nicolas and Nessau, Hendrik and Reke, Michael and Ferrein, Alexander}, title = {Model-predictive control with parallelised optimisation for the navigation of autonomous mining vehicles}, series = {2023 IEEE Intelligent Vehicles Symposium (IV)}, booktitle = {2023 IEEE Intelligent Vehicles Symposium (IV)}, publisher = {IEEE}, isbn = {979-8-3503-4691-6 (Online)}, doi = {10.1109/IV55152.2023.10186806}, pages = {6 Seiten}, year = {2023}, abstract = {The work in modern open-pit and underground mines requires the transportation of large amounts of resources between fixed points. The navigation to these fixed points is a repetitive task that can be automated. The challenge in automating the navigation of vehicles commonly used in mines is the systemic properties of such vehicles. Many mining vehicles, such as the one we have used in the research for this paper, use steering systems with an articulated joint bending the vehicle's drive axis to change its course and a hydraulic drive system to actuate axial drive components or the movements of tippers if available. To address the difficulties of controlling such a vehicle, we present a model-predictive approach for controlling the vehicle. While the control optimisation based on a parallel error minimisation of the predicted state has already been established in the past, we provide insight into the design and implementation of an MPC for an articulated mining vehicle and show the results of real-world experiments in an open-pit mine environment.}, language = {en} } @inproceedings{EichenbaumNikolovskiMuelhensetal.2023, author = {Eichenbaum, Julian and Nikolovski, Gjorgji and M{\"u}lhens, Leon and Reke, Michael and Ferrein, Alexander and Scholl, Ingrid}, title = {Towards a lifelong mapping approach using Lanelet 2 for autonomous open-pit mine operations}, series = {2023 IEEE 19th International Conference on Automation Science and Engineering (CASE)}, booktitle = {2023 IEEE 19th International Conference on Automation Science and Engineering (CASE)}, publisher = {IEEE}, isbn = {979-8-3503-2069-5 (Online)}, doi = {10.1109/CASE56687.2023.10260526}, pages = {8 Seiten}, year = {2023}, abstract = {Autonomous agents require rich environment models for fulfilling their missions. High-definition maps are a well-established map format which allows for representing semantic information besides the usual geometric information of the environment. These are, for instance, road shapes, road markings, traffic signs or barriers. The geometric resolution of HD maps can be as precise as of centimetre level. In this paper, we report on our approach of using HD maps as a map representation for autonomous load-haul-dump vehicles in open-pit mining operations. As the mine undergoes constant change, we also need to constantly update the map. Therefore, we follow a lifelong mapping approach for updating the HD maps based on camera-based object detection and GPS data. We show our mapping algorithm based on the Lanelet 2 map format and show our integration with the navigation stack of the Robot Operating System. We present experimental results on our lifelong mapping approach from a real open-pit mine.}, language = {en} }