Refine
Year of publication
Document Type
- Conference Proceeding (8)
- Article (3)
- Doctoral Thesis (1)
Has Fulltext
- no (12) (remove)
Keywords
- 3D object detection (1)
- Building Automation (1)
- CCD-Bildwandler (1)
- CO2 (1)
- Carbon Dioxide (1)
- Deep learning (1)
- Dreidimensionale Bildverarbeitung (1)
- Education (1)
- Klassifikator <Informatik> (1)
- LiDAR (1)
- Objekterkennung (1)
- Smart Building (1)
- Three-dimensional displays (1)
- Vorverarbeitung (1)
- autonomous driving (1)
- do-it-yourself (1)
- education (1)
- embedded hardware (1)
- information systems (1)
- sensor networks (1)
Institute
- ECSM European Center for Sustainable Mobility (12) (remove)
Existing residential buildings have an average lifetime of 100 years. Many of these buildings will exist for at least another 50 years. To increase the efficiency of these buildings while keeping costs at reasonable rates, they can be retrofitted with sensors that deliver information to central control units for heating, ventilation and electricity. This retrofitting process should happen with minimal intervention into existing infrastructure and requires new approaches for sensor design and data transmission. At FH Aachen University of Applied Sciences, students of different disciplines work together to learn how to design, build, deploy and operate such sensors. The presented teaching project already created a low power design for a combined CO2, temperature and humidity measurement device that can be easily integrated into most home automation systems
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.