Refine
Departments, institutes and facilities
- Fachbereich Informatik (979)
- Fachbereich Angewandte Naturwissenschaften (630)
- Institut für funktionale Gen-Analytik (IFGA) (560)
- Institut für Technik, Ressourcenschonung und Energieeffizienz (TREE) (393)
- Fachbereich Ingenieurwissenschaften und Kommunikation (386)
- Fachbereich Wirtschaftswissenschaften (319)
- Institute of Visual Computing (IVC) (286)
- Institut für Cyber Security & Privacy (ICSP) (244)
- Institut für Verbraucherinformatik (IVI) (166)
- Internationales Zentrum für Nachhaltige Entwicklung (IZNE) (126)
- Fachbereich Sozialpolitik und Soziale Sicherung (68)
- Institut für Sicherheitsforschung (ISF) (66)
- Graduierteninstitut (51)
- Zentrum für Innovation und Entwicklung in der Lehre (ZIEL) (21)
- Sprachenzentrum (19)
- Institut für Detektionstechnologien (IDT) (15)
- Centrum für Entrepreneurship, Innovation und Mittelstand (CENTIM) (13)
- Präsidium (11)
- Stabsstelle Kommunikation und Marketing (11)
- Institut für KI und Autonome Systeme (A2S) (7)
- Zentrum für Ethik und Verantwortung (ZEV) (7)
- Verwaltung (4)
- Zentrum für Wissenschafts- und Technologietransfer (ZWT) (4)
- Institut für IT-Service (ITS) (2)
- Bibliothek (1)
- Institut für Soziale Innovationen (ISI) (1)
Document Type
- Article (1649)
- Conference Object (1426)
- Part of a Book (249)
- Preprint (88)
- Report (73)
- Doctoral Thesis (64)
- Book (monograph, edited volume) (57)
- Master's Thesis (36)
- Working Paper (34)
- Conference Proceedings (27)
- Research Data (23)
- Lecture (18)
- Part of Periodical (14)
- Contribution to a Periodical (9)
- Patent (8)
- Bachelor Thesis (5)
- Book review (4)
- Other (3)
- Study Thesis (3)
- Diploma Thesis (1)
- Postdoctoral thesis (1)
Year of publication
Language
- English (3792) (remove)
Keywords
- Machine Learning (15)
- Virtual Reality (15)
- FPGA (14)
- Robotics (14)
- Sustainability (14)
- ENaC (13)
- GC/MS (13)
- virtual reality (13)
- sustainability (12)
- ICT (11)
There is severe clinical vitamin A deficiency (VAD) prevalence among Ghanaians and many African countries. Foodbased diets has been suggested as a more sustainable approach to solving the VAD situation in Africa. In this study, A participatory action research between orange flesh sweet potato farmers, gari processors within central region and academia was adopted to develop gari containing provitamin A beta-carotene. Gari is a major staple for Ghanaians and people in the West African subregion due to its affordability and swelling capacity. It is mainly eaten raw with water, sugar, groundnut and milk as gari-soakings or with hot water to prepare gelatinized food called gari-kai in Ghana or “eba” among Nigerians. However, gari is limited in provitamin A carotenoids. Orange fleshed sweet potato (OFSP) is known to contain large amount of vitamin A precursor. Therefore, addition of OFSP to gari would have the potential to fight the high prevalence rate of vitamin A deficiency amongst less developed regions of Africa. To develop this, different proportions of orange fleshed sweet potatoes (OFSP) was used to substitute cassava mash and fermented spontaneously to produce composite gari - a gritty-crispy ready-to-eat food product. Both the amount of OFSP and the fermentation duration caused significant increases in the β-carotene content of the composite gari. OFSP addition reduced the luminance while roasting made the composite gari yellower when compared with the cake used. Addition of OFSP negatively affected the swelling capacity of the gari although not significant. The taste, texture, flavour and the overall preferences for the composite gari decreased due to the addition of the OFSP but fermentation duration (FD) improved them. The sample with 10% OFSP and FD of 1.81 days was found to produce the optimal gari. One-portion of the optimal gari would contribute to 34.75, 23.2, 23.2, 27, 17 and 16% of vitamin A requirements amongst children, adolescent, adult males, adult females, pregnant women and lactating mothers respectively. The study demonstrated that partial substitution of cassava with OFSP for gari production would have the potential to fight the high prevalence rate of vitamin A deficiency amongst less developed regions of Africa while involvement of farmers and processors prior to the design of research phase enhanced the adoption of intervention strategies.
Neuromorphic computing aims to mimic the computational principles of the brain in silico and has motivated research into event-based vision and spiking neural networks (SNNs). Event cameras (ECs) capture local, independent changes in brightness, and offer superior power consumption, response latencies, and dynamic ranges compared to frame-based cameras. SNNs replicate neuronal dynamics observed in biological neurons and propagate information in sparse sequences of ”spikes”. Apart from biological fidelity, SNNs have demonstrated potential as an alternative to conventional artificial neural networks (ANNs), such as in reducing energy expenditure and inference time in visual classification. Although potentially beneficial for robotics, the novel event-driven and spike-based paradigms remain scarcely explored outside the domain of aerial robots.
To investigate the utility of brain-inspired sensing and data processing in a robotics application, we developed a neuromorphic approach to real-time, online obstacle avoidance on a manipulator with an onboard camera. Our approach adapts high-level trajectory plans with reactive maneuvers by processing emulated event data in a convolutional SNN, decoding neural activations into avoidance motions, and adjusting plans in a dynamic motion primitive formulation. We conducted simulated and real experiments with a Kinova Gen3 arm performing simple reaching tasks involving static and dynamic obstacles. Our implementation was systematically tuned, validated, and tested in sets of distinct task scenarios, and compared to a non-adaptive baseline through formalized quantitative metrics and qualitative criteria.
The neuromorphic implementation facilitated reliable avoidance of imminent collisions in most scenarios, with 84% and 92% median success rates in simulated and real experiments, where the baseline consistently failed. Adapted trajectories were qualitatively similar to baseline trajectories, indicating low impacts on safety, predictability and smoothness criteria. Among notable properties of the SNN were the correlation of processing time with the magnitude of perceived motions (captured in events) and robustness to different event emulation methods. Preliminary tests with a DAVIS346 EC showed similar performance, validating our experimental event emulation method. These results motivate future efforts to incorporate SNN learning, utilize neuromorphic processors, and target other robot tasks to further explore this approach.
An essential measure of autonomy in service robots designed to assist humans is adaptivity to the various contexts of human-oriented tasks. These robots may have to frequently execute the same action, but subject to subtle variations in task parameters that determine optimal behaviour. Such actions are traditionally executed by robots using pre-determined, generic motions, but a better approach could utilize robot arm maneuverability to learn and execute different trajectories that work best in each context.
In this project, we explore a robot skill acquisition procedure that allows incorporating contextual knowledge, adjusting executions according to context, and improvement through experience, as a step towards more adaptive service robots. We propose an apprenticeship learning approach to achieving context-aware action generalisation on the task of robot-to-human object hand-over. The procedure combines learning from demonstration, with which a robot learns to imitate a demonstrator’s execution of the task, and a reinforcement learning strategy, which enables subsequent experiential learning of contextualized policies, guided by information about context that is integrated into the learning process. By extending the initial, static hand-over policy to a contextually adaptive one, the robot derives and executes variants of the demonstrated action that most appropriately suit the current context. We use dynamic movement primitives (DMPs) as compact motion representations, and a model-based Contextual Relative Entropy Policy Search (C-REPS) algorithm for learning policies that can specify hand-over position, trajectory shape, and execution speed, conditioned on context variables. Policies are learned using simulated task executions, before transferring them to the robot and evaluating emergent behaviours.
We demonstrate the algorithm’s ability to learn context-dependent hand-over positions, and new trajectories, guided by suitable reward functions, and show that the current DMP implementation limits learning context-dependent execution speeds. We additionally conduct a user study involving participants assuming different postures and receiving an object from the robot, which executes hand-overs by either exclusively imitating a demonstrated motion, or selecting hand-over positions based on learned contextual policies and adapting its motion accordingly. The results confirm the hypothesized improvements in the robot’s perceived behaviour when it is context-aware and adaptive, and provide useful insights that can inform future developments.
An essential measure of autonomy in assistive service robots is adaptivity to the various contexts of human-oriented tasks, which are subject to subtle variations in task parameters that determine optimal behaviour. In this work, we propose an apprenticeship learning approach to achieving context-aware action generalization on the task of robot-to-human object hand-over. The procedure combines learning from demonstration and reinforcement learning: a robot first imitates a demonstrator’s execution of the task and then learns contextualized variants of the demonstrated action through experience. We use dynamic movement primitives as compact motion representations, and a model-based C-REPS algorithm for learning policies that can specify hand-over position, conditioned on context variables. Policies are learned using simulated task executions, before transferring them to the robot and evaluating emergent behaviours. We additionally conduct a user study involving participants assuming different postures and receiving an object from a robot, which executes hand-overs by either imitating a demonstrated motion, or adapting its motion to hand-over positions suggested by the learned policy. The results confirm the hypothesized improvements in the robot’s perceived behaviour when it is context-aware and adaptive, and provide useful insights that can inform future developments.
Machine learning-based solutions are frequently adapted in several applications that require big data in operations. The performance of a model that is deployed into operations is subject to degradation due to unanticipated changes in the flow of input data. Hence, monitoring data drift becomes essential to maintain the model’s desired performance. Based on the conducted review of the literature on drift detection, statistical hypothesis testing enables to investigate whether incoming data is drifting from training data. Because Maximum Mean Discrepancy (MMD) and Kolmogorov-Smirnov (KS) have shown to be reliable distance measures between multivariate distributions in the literature review, both were selected from several existing techniques for experimentation. For the scope of this work, the image classification use case was experimented with using the Stream-51 dataset. Based on the results from different drift experiments, both MMD and KS showed high Area Under Curve values. However, KS exhibited faster performance than MMD with fewer false positives. Furthermore, the results showed that using the pre-trained ResNet-18 for feature extraction maintained the high performance of the experimented drift detectors. Furthermore, the results showed that the performance of the drift detectors highly depends on the sample sizes of the reference (training) data and the test data that flow into the pipeline’s monitor. Finally, the results also showed that if the test data is a mixture of drifting and non-drifting data, the performance of the drift detectors does not depend on how the drifting data are scattered with the non-drifting ones, but rather their amount in the test set
In this paper, the performance evaluation of Frequency Modulated Chaotic On-Off Keying (FM-COOK) in AWGN, Rayleigh and Rician fading channels is given. The simulation results show that an improvement in BER can be gained by incorporating the FM modulation with COOK for SNR values less than 10dB in AWGN case and less than 6dB for Rayleigh and Rician fading channels.
ICT integration by universities teaching professionals is emerging as a major concern, this study demonstrate the need to address the integration problem by encouraging existing metrics use in indexing ICT integration as an ICT governance strategy. Quality of integration depends on quality indexing which also depend on quality of existing metrics and their use. Considering the role that University Information Technology Teaching Professionals’ (UITTPs) continuous improvement indexing can offer, towards autonomic governance of the continuous emerging ICTs in the university teaching, this study examined extent in use of existing ICT integration metrics to index ICT integration by the UITTPs. Six metrics for ICT integration were investigated; time, workshop course content relevance, technical malfunctions, support conditions, support services, and motivation and commitment to student learning and staff professional development metrics. Descriptive survey design was used in which interviews were conducted to UITTPs in three (3) public and three (3) private purposively selected universities in Kenya. The findings were analyzed descriptively and inferentially using Kendall’s correlation of concordance and tested using Chi-square on the extent of concordance and presented with help of frequency tables, figures and percentages. The findings revealed that all the metrics are rarely used for indexing ICT integration (32.8%), and most UITTPs were in discordance on this level of all the six metrics use except for support condition. This implied that the use of metrics for indexing integration has not been formalized across the Kenyan universities. Universities need to be encouraged to identify suitable metrics, formalize them and improve their frequency in use. Secondly, socio based metrics such as content relevance are used more frequently for indexing integration as compared to Technical metrics, socio-technical metrics balance therefore need to be emphasized by the universities management when determining and using metrics for indexing ICT integration.
The design of an efficient digital circuit in term of low-power has become a very challenging issue. For this reason, low-power digital circuit design is a topic addressed in electrical and computer engineering curricula, but it also requires practical experiments in a laboratory. This PhD research investigates a novel approach, the low-power design laboratory system by developing a new technical and pedagogical system. The low-power design laboratory system is composed of two types of laboratories: the on-site (hands-on) laboratory and the remote laboratory. It has been developed at the Bonn-Rhine-Sieg University of Applied Sciences to teach low-power techniques in the laboratory. Additionally, this thesis contributes a suggestion on how the learning objectives can be complemented by developing a remote system in order to improve the teaching process of the low-power digital circuit design. This laboratory system enables online experiments that can be performed using physical instruments and obtaining real data via the internet. The laboratory experiments use a Field Programmable Gate Array (FPGA) as a design platform for circuit implementation by students and use image processing as an application for teaching low-power techniques.
This thesis presents the instructions for the low-power design experiments which use a top-down hierarchical design methodology. The engineering student designs his/her algorithm with a high level of abstraction and the experimental results are obtained and measured at a low level (hardware) so that more information is available to correctly estimate the power dissipation such as specification, latency, thermal effect, and technology used. Power dissipation of the digital system is influenced by specification, design, technology used, as well as operating temperature. Digital circuit designers can observe the most influential factors in power dissipation during the laboratory exercises in the on-site system and then use the remote system to supplement investigating the other factors. Furthermore, the remote system has obvious benefits such as developing learning outcomes, facilitating new teaching methods, reducing costs and maintenance, cost-saving by reducing the numbers of instructors, saving instructor time and simplifying their tasks, facilitating equipment sharing, improving reliability, and finally providing flexibility of usage the laboratories.
For many different applications, current information about the bandwidth-related metrics of the utilized connection is very useful as they directly impact the performance of throughput sensitive applications such as streaming servers, IPTV and VoIP applications. In literature, several tools have been proposed to estimate major bandwidth-related metrics such as capacity, available bandwidth and achievable throughput. The vast majority of these tools fall into one of Packet Pair (PP), Variable Packet Size (VPS), Self-Loading of Periodic Streams (SLoPS) or Throughput approaches. In this study, seven popular bandwidth estimation tools including nettimer, pathrate, pathchar, pchar, clink, pathload and iperf belonging to these four well-known estimation techniques are presented and experimentally evaluated in a controlled testbed environment. Differently from the rest of studies in literature, all tools have been uniformly classified and evaluated according to an objective and sophisticated classification and evaluation scheme. The performance comparison of the tools incorporates not only the estimation accuracy but also the probing time and overhead caused.
YAWL (Yet Another Workflow Language) is an open source Business Process Management System, first released in 2003. YAWL grew out of a university research environment to become a unique system that has been deployed worldwide as a laboratory environment for research in Business Process Management and as a productive system in other scientific domains.
Cancer is one of the leading causes of death worldwide [183], with lung tumors being the most frequent cause of cancer deaths in men as well as one of the most common cancers diagnosed in woman [40]. As symptoms often arise in advanced stages, an early diagnosis is especially important to ensure the best and earliest possible treatment. In order to achieve this, Computed Tomography (CT) scans are frequently used for tumor detection and diagnosis. We will present examples of publicly available CT image data of lung cancer patients and discuss possible methods to realize an automatic system for automated cancer diagnosis. We will also look at the recent SPIE-AAPM Lung CT Challenge [10] data set in detail and describe possible methods and challenges for image segmentation and classification based on this data set.
With trainings and research oriented towards sustainable development since 2006 (Water and Sanitation, Infrastructure, Renewable Energies and Energy Processes), Foundation 2iE is positioning itself as a reference institute that trains innovative engineers-entrepreneurs for the needs and challenges of Africa’s development. Center of Excellence of the UEMOA and the World Bank, CSR is at the heart of the Strategy of the institute which aims to be a showcase in this field in Africa.
When users in virtual reality cannot physically walk and self-motions are instead only visually simulated, spatial updating is often impaired. In this paper, we report on a study that investigated if HeadJoystick, an embodied leaning-based flying interface, could improve performance in a 3D navigational search task that relies on maintaining situational awareness and spatial updating in VR. We compared it to Gamepad, a standard flying interface. For both interfaces, participants were seated on a swivel chair and controlled simulated rotations by physically rotating. They either leaned (forward/backward, right/left, up/down) or used the Gamepad thumbsticks for simulated translation. In a gamified 3D navigational search task, participants had to find eight balls within 5 min. Those balls were hidden amongst 16 randomly positioned boxes in a dark environment devoid of any landmarks. Compared to the Gamepad, participants collected more balls using the HeadJoystick. It also minimized the distance travelled, motion sickness, and mental task demand. Moreover, the HeadJoystick was rated better in terms of ease of use, controllability, learnability, overall usability, and self-motion perception. However, participants rated HeadJoystick could be more physically fatiguing after a long use. Overall, participants felt more engaged with HeadJoystick, enjoyed it more, and preferred it. Together, this provides evidence that leaning-based interfaces like HeadJoystick can provide an affordable and effective alternative for flying in VR and potentially telepresence drones.
Queueing Theory
(2024)
The nature of the program was an exchange program between Hochschule Bonn-Rhein-Sieg, University of Applied Sciences and the University of Cape Coast. The program was advertised and we applied. We were shortlisted for interview and we were selected as the candidates for the exchange program. The program took a period of five months. We set off from Accra, Ghana to Germany on 7th September 2015, and returned to Ghana on 25th January 2016.
The role of tourism entrepreneurship in rural development continues to be a subject of interest and debate among academia and practitioners. Theoretically, it is anticipated that tourism entrepreneurship will lead to livelihood diversification, enhancement and ultimately a revitalization of the rural economy. While tourism is posited as an accessible entrepreneurship pathway, there is a dearth of information regarding rural dwellers’ actual experiences with it, especially within the Ghanaian context. Using a case study approach and qualitative data from Wli; a rural tourism destination in Ghana, this paper delves into the opportunities and concerns associated with tourism entrepreneurship in rural areas. Data was obtained between November and December 2016 from 27 persons who were either tourism enterprise owners or employees. Findings from the study showed that entrepreneurial activities centred on the provision of accommodation, food and beverage, souvenir and guiding services. The nature of the activities enabled easy transfer of existing skills and knowledge. Further, entry into tourism entrepreneurship was perceived to be easy by the majority of study participants. These findings confirm the potential for tourism to be employed in boosting entrepreneurial activities in rural areas. Nevertheless, there were concerns regarding access to credit, institutional support, unhealthy competitions, low incomes, unguaranteed pensions, and seasonality and skewness of demand. These concerns threatened the growth and sustainability of tourism entrepreneurship within the community. From a policy perspective, there is a need for institutional recognition and support for tourism entrepreneurial intentions and activities in rural areas. Practice-wise, credit facilities need to be designed specifically for tourism-related rural enterprises. Further, periodic skills and knowledge augmentation programmes must be initiated to help expand the skill sets for the rural entrepreneurs. Finally, there is a need for the formation of traderelated networks to provide a platform for knowledge and experience sharing among the entrepreneurs.
Entrepreneurship education serves a conduit for new venture creation as it provides the knowledge and skills needed to increase the self-efficacy of individuals to start and run new businesses and to grow existing ones. This study, therefore, sought to assess the relationship between the approaches to the teaching of entrepreneur-ship and entrepreneurial intention on a cohort of 292 respondents consisting of students who have studied entrepreneurship in three selected Universities. A structured questionnaire was used to obtain data randomly from students. The canonical correlation results indicate that education for and through entrepreneurship is the best approach to promoting entrepreneurial intensity among University students, if the aim of teaching entrepreneur-ship is to promote start-up activities. The findings provide valuable insights for institutions of higher learning and policy makers in Ghana with respect to the appropriate methodologies to be adopted in the teaching of entrepreneurship in our universities.
In the last two decades, studies that analyse the political economy of sustainable energy transitions have increasingly become available. Yet very few attempts have been made to synthesize the factors discussed in the growing literature. This paper reviews the extant empirical literature on the political economy of sustainable energy transitions. Using a well-defined search strategy, a total of 36 empirical contributions covering the period 2008 to 2022 are reviewed full text. Overall, the findings highlight the role of vested interest, advocacy coalitions and green constituencies, path dependency, external shocks, policy and institutional environment, political institutions and fossil fuel resource endowments as major political economy factors influencing sustainable energy transitions across both high income countries, and low and middle income countries. In addition, the paper highlights and discusses some critical knowledge gaps in the existing literature and provides suggestions for a future research agenda.
As competition for tourists becomes more global, understanding and accommodating the needs of international tourists, with their different cultural backgrounds, has become increasingly important. This study highlights the variations in tourist industry service--particularly as they relate to different cultures. Specifically, service failures experienced by Japanese and German tourists in the U.S. were categorized using the Critical Incident Technique (CIT). The results were compared with earlier studies of service failures experienced by American consumers in the tourist industry. The sample consists of 128 Japanese and 94 “Germanic” (German, Austrian, Swiss-German) respondents. The Japanese and German sample rated “Inappropriate employee behavior” most significant category of service failure. More than half of these respondents said that, because of the failure, they would avoid the offending U.S. business. This is a much stronger response than an American sample had reported in an earlier study. The implications for managers and researchers are discussed.
Robust Indoor Localization Using Optimal Fusion Filter For Sensors And Map Layout Information
(2014)
Competency-Based Teaching Using Simulation Exercises: Evidence of the University of Cape Coast
(2018)
Tertiary institutions exist to train manpower to solve local, national, and international problems. Products from such institutions should not be a problem to countries as in the case of some Sub-Saharan African countries including Ghana which has a high level of graduate unemployment. Among the causes of the problem is the nature of teaching or the syllabus or the programs students pursue while in such institutions. The paper discusses one of the teaching strategies used to make a course relevant for a program and for the working world. In this course, students are introduced to practice-oriented learning through simulation exercises. The project activities specifically seek to assess the students’ understanding of business formation; examine students’ understanding of sustainability, creativity and innovation of business ideas; assess their understanding of the functional areas of business including marketing & sales, finance, human resource management, operations, and accounting, among others. Feedback from students who have participated indicates the exercise gave much more exposure and meaning to the concepts they learned in class. In this exercise, students build teams, develop a product, learn to set up a business, and design organogram, business vision, mission, and core values. The exercise empowers students to learn by doing. It accords students the opportunity to review their own knowledge and skills with respect to the concepts they have learned in the course. More than 3000 students have participated in this project since its inception in the academic year 2013/2014. It is estimated that 1000 students will participate in this project in the academic year 2017/2018.
Conclusion
(2018)
There is a paradigm shift from traditional content-based education and training to competencybased and practice-oriented training. This shift has occurred because practice-oriented teaching has been found to produce a training outcome that is industry focused, generating the relevant occupational standards. Competency-based training program often comprises of modules broken into segments called learning outcomes. These learning outcomes are based on criteria set by industry and assessment is designed to ensure students become competent in their respective areas of specialization.
Multidisciplinary, multicultural, and multitasking has taken center stage in the global educational debate. Globalization and improvement in communication have affected the way organisations operate and hence influenced whom they hire. Today, it is common practice to work with people from diverse backgrounds and it requires competencies that go beyond general project management. Intercultural awareness, networking in different global communities, and learning to develop specific communication strategies for different stakeholders is all part of the package of skills and competencies that are required in today's interconnected world. This has indirect implication on the nature of skills and competencies institutions/universities must equip their students with to enable them to compete successfully in the working world.
Airborne and spaceborne platforms are the primary data sources for large-scale forest mapping, but visual interpretation for individual species determination is labor-intensive. Hence, various studies focusing on forests have investigated the benefits of multiple sensors for automated tree species classification. However, transferable deep learning approaches for large-scale applications are still lacking. This gap motivated us to create a novel dataset for tree species classification in central Europe based on multi-sensor data from aerial, Sentinel-1 and Sentinel-2 imagery. In this paper, we introduce the TreeSatAI Benchmark Archive, which contains labels of 20 European tree species (i.e., 15 tree genera) derived from forest administration data of the federal state of Lower Saxony, Germany. We propose models and guidelines for the application of the latest machine learning techniques for the task of tree species classification with multi-label data. Finally, we provide various benchmark experiments showcasing the information which can be derived from the different sensors including artificial neural networks and tree-based machine learning methods. We found that residual neural networks (ResNet) perform sufficiently well with weighted precision scores up to 79 % only by using the RGB bands of aerial imagery. This result indicates that the spatial content present within the 0.2 m resolution data is very informative for tree species classification. With the incorporation of Sentinel-1 and Sentinel-2 imagery, performance improved marginally. However, the sole use of Sentinel-2 still allows for weighted precision scores of up to 74 % using either multi-layer perceptron (MLP) or Light Gradient Boosting Machine (LightGBM) models. Since the dataset is derived from real-world reference data, it contains high class imbalances. We found that this dataset attribute negatively affects the models' performances for many of the underrepresented classes (i.e., scarce tree species). However, the class-wise precision of the best-performing late fusion model still reached values ranging from 54 % (Acer) to 88 % (Pinus). Based on our results, we conclude that deep learning techniques using aerial imagery could considerably support forestry administration in the provision of large-scale tree species maps at a very high resolution to plan for challenges driven by global environmental change. The original dataset used in this paper is shared via Zenodo (https://doi.org/10.5281/zenodo.6598390, Schulz et al., 2022). For citation of the dataset, we refer to this article.
This paper presents the b-it-bots RoboCup@Work team and its current hardware and functional architecture for the KUKA youBot robot.We describe the underlying software framework and the developed capabilities required for operating in industrial environments including features such as reliable and precise navigation, flexible manipulation and robust object recognition.