Accurate localization in autonomous robots enables effective decision-making within their operating environment. Various methods have been developed to address this challenge, encompassing traditional techniques, fiducial marker utilization, and machine learning approaches. This work proposes a deep-learning solution employing Convolutional Neural Networks (CNN) to tackle the localization problem, specifically in the context of the RobotAtFactory 4.0 competition. The proposed approach leverages transfer learning from the pre-trained VGG16 model to capitalize on its existing knowledge. To validate the effectiveness of the approach, a simulated scenario was employed. The experimental results demonstrated an error within the millimeter scale and rapid response times in milliseconds. Notably, the presented approach offers several advantages, including a consistent model size regardless of the number of training images utilized and the elimination of the need to know the absolute positions of the fiducial markers.
DOCUMENT
There is an increasing interest in how to create an effective and comfortable indoor environment for lecturers and students in higher education. To achieve evidence-based improvements in the indoor environmental quality (IEQ) of higher education learning environments, this research aimed to gain new knowledge for creating optimal indoor environmental conditions that best facilitate in-class activities, i.e. teaching and learning, and foster academic achievement. The academic performance of lecturers and students is subdivided into short-term academic performance, for example, during a lecture and long-term academic performance, during an academic course or year, for example. First, a systematic literature review was conducted to reveal the effect of indoor environmental quality in classrooms in higher education on the quality of teaching, the quality of learning, and students’ academic achievement. With the information gathered on the applied methods during the literature review, a systematic approach was developed and validated to capture the effect of the IEQ on the main outcomes. This approach enables research that aims to examine the effect of all four IEQ parameters, indoor air quality, thermal conditions, lighting conditions, and acoustic conditions on students’ perceptions, responses, and short-term academic performance in the context of higher education classrooms. Next, a field experiment was conducted, applying the validated systematic approach, to explore the effect of multiple indoor environmental parameters on students and their short-term academic performance in higher education. Finally, a qualitative case study gathered lecturers’ and students’ perceptions related to the IEQ. Furthermore, how these users interact with the environment to maintain an acceptable IEQ was studied. During the systematic literature review, multiple scientific databases were searched to identify relevant scientific evidence. After the screening process, 21 publications were included. The collected evidence showed that IEQ can contribute positively to students’ academic achievement. However, it can also affect the performance of students negatively, even if the IEQ meets current standards for classrooms’ IEQ conditions. Not one optimal IEQ was identified after studying the evidence. Indoor environmental conditions in which students perform at their best differ and are task depended, indicating that classrooms should facilitate multiple indoor environmental conditions. Furthermore, the evidence provides practical information for improving the design of experimental studies, helps researchers in identifying relevant parameters, and lists methods to examine the influence of the IEQ on users. The measurement methods deduced from the included studies of the literature review, were used for the development of a systematic approach measuring classroom IEQ and students’ perceived IEQ, internal responses, and short-term academic performance. This approach allowed studying the effect of multiple IEQ parameters simultaneously and was tested in a pilot study during a regular academic course. The perceptions, internal responses, and short-term academic performance of participating students were measured. The results show associations between natural variations of the IEQ and students’ perceptions. These perceptions were associated with their physiological and cognitive responses. Furthermore, students’ perceived cognitive responses were associated with their short-term academic performance. These observed associations confirm the construct validity of the composed systematic approach. S Summary 9 This systematic approach was then applied in a field experiment, to explore the effect of multiple indoor environmental parameters on students and their short-term academic performance in higher education. A field study, with a between-groups experimental design, was conducted during a regular academic course in 2020-2021 to analyze the effect of different acoustic, lighting, and indoor air quality (IAQ) conditions. First, the reverberation time was manipulated to 0.4 s in the intervention condition (control condition 0.6 s). Second, the horizontal illuminance level was raised from 500 to 750 lx in the intervention condition (control condition 500 lx). These conditions correspond with quality class A (intervention condition) and B (control condition), specified in Dutch IEQ guidelines for school buildings (2015). Third, the IAQ, which was ~1100 ppm carbon dioxide (CO2), as a proxy for IAQ, was improved to CO2 concentrations under 800 ppm, meeting quality class A in both conditions. Students’ perceptions were measured during seven campaigns with a questionnaire; their actual cognitive and shortterm academic performances were evaluated with validated tests and an academic test, composed by the lecturer, as a subject-matter-expert on the taught topic, covered subjects discussed during the lecture. From 201 students 527 responses were collected and analyzed. A reduced RT in combination with raised HI improved students’ perceptions of the lighting environment, internal responses, and quality of learning. However, this experimental condition negatively influenced students’ ability to solve problems, while students' content-related test scores were not influenced. This shows that although quality class A conditions for RT and HI improved students’ perceptions, it did not influence their short-term academic performance. Furthermore, the benefits of reduced RT in combination with raised HI were not observed in improved IAQ conditions. Whether the sequential order of the experimental conditions is relevant in inducing these effects and/or whether improving two parameters is already beneficial, is unknown Finally, a qualitative case study explored lecturers’ and students’ perceptions of the IEQ of classrooms, which are suitable to give tutorials with a maximum capacity of about 30 students. Furthermore, how lecturers and students interact with this indoor environment to maintain an acceptable IEQ was examined. Eleven lecturers of the Hanze University of Applied Sciences (UAS), located in the northern part of the Netherlands, and twenty-four of its students participated in three focus group discussions. The findings show that lecturers and students experience poor thermal, lighting, acoustic, and IAQ conditions which may influence teaching and learning performance. Furthermore, maintaining acceptable thermal and IAQ conditions was difficult for lecturers as opening windows or doors caused noise disturbances. In uncomfortable conditions, lecturers may decide to pause earlier or shorten a lecture. When students experienced discomfort, it may affect their ability to concentrate, their emotional status, and their quality of learning. Acceptable air and thermal conditions in classrooms will mitigate the need to open windows and doors. This allows lecturers to keep doors and windows closed, combining better classroom conditions with neither noise disturbances nor related distractions. Designers and engineers should take these end users’ perceptions into account, often monitored by facility management (FM), during the renovation or construction of university buildings to achieve optimal IEQ conditions in higher education classrooms. Summary 10 The results of these four studies indicate that there is not a one-size fits all indoor environmental quality to facilitate optimal in-class activities. Classrooms’ thermal environment should be effectively controlled with the option of a local (manual) intervention. Classrooms’ lighting conditions should also be adjustable, both in light color and light intensity. This enables lecturers to adjust the indoor environment to facilitate in-class activities optimally. Lecturers must be informed by the building operator, for example, professionals of the Facility Department, how to change classrooms’ IEQ settings. And this may differ per classroom because each building, in which the classroom is located, is operated differently apart from the classroom location in the building, exposure to the environment, and its use. The knowledge that has come available from this study, shows that optimal indoor environmental conditions can positively influence lecturers’ and students’ comfort, health, emotional balance, and performance. These outcomes have the capacity to contribute to an improved school climate and thus academic achievement.
LINK
The use of machine learning in embedded systems is an interesting topic, especially with the growth in popularity of the Internet of Things (IoT). The capacity of a system, such as a robot, to self-localize, is a fundamental skill for its navigation and decision-making processes. This work focuses on the feasibility of using machine learning in a Raspberry Pi 4 Model B, solving the localization problem using images and fiducial markers (ArUco markers) in the context of the RobotAtFactory 4.0 competition. The approaches were validated using a realistically simulated scenario. Three algorithms were tested, and all were shown to be a good solution for a limited amount of data. Results also show that when the amount of data grows, only Multi-Layer Perception (MLP) is feasible for the embedded application due to the required training time and the resulting size of the model.
DOCUMENT
Drones have been verified as the camera of 2024 due to the enormous exponential growth in terms of the relevant technologies and applications such as smart agriculture, transportation, inspection, logistics, surveillance and interaction. Therefore, the commercial solutions to deploy drones in different working places have become a crucial demand for companies. Warehouses are one of the most promising industrial domains to utilize drones to automate different operations such as inventory scanning, goods transportation to the delivery lines, area monitoring on demand and so on. On the other hands, deploying drones (or even mobile robots) in such challenging environment needs to enable accurate state estimation in terms of position and orientation to allow autonomous navigation. This is because GPS signals are not available in warehouses due to the obstruction by the closed-sky areas and the signal deflection by structures. Vision-based positioning systems are the most promising techniques to achieve reliable position estimation in indoor environments. This is because of using low-cost sensors (cameras), the utilization of dense environmental features and the possibilities to operate in indoor/outdoor areas. Therefore, this proposal aims to address a crucial question for industrial applications with our industrial partners to explore limitations and develop solutions towards robust state estimation of drones in challenging environments such as warehouses and greenhouses. The results of this project will be used as the baseline to develop other navigation technologies towards full autonomous deployment of drones such as mapping, localization, docking and maneuvering to safely deploy drones in GPS-denied areas.
Due to the exponential growth of ecommerce, the need for automated Inventory management is crucial to have, among others, up-to-date information. There have been recent developments in using drones equipped with RGB cameras for scanning and counting inventories in warehouse. Due to their unlimited reach, agility and speed, drones can speed up the inventory process and keep it actual. To benefit from this drone technology, warehouse owners and inventory service providers are actively exploring ways for maximizing the utilization of this technology through extending its capability in long-term autonomy, collaboration and operation in night and weekends. This feasibility study is aimed at investigating the possibility of developing a robust, reliable and resilient group of aerial robots with long-term autonomy as part of effectively automating warehouse inventory system to have competitive advantage in highly dynamic and competitive market. To that end, the main research question is, “Which technologies need to be further developed to enable collaborative drones with long-term autonomy to conduct warehouse inventory at night and in the weekends?” This research focusses on user requirement analysis, complete system architecting including functional decomposition, concept development, technology selection, proof-of-concept demonstrator development and compiling a follow-up projects.
