Examensarbeten för masterexamen
Länka till denna samling:
Browse
Senast publicerade
Visar 1 - 5 av 2019
- PostUncertainty estimation in multi-modal 3D object detection(2024) Rosén, Anton; Chalmers tekniska högskola / Institutionen för elektroteknik; Hammarstrand, Lars; Johnander, Joakim; Fatemi, Maryam; Lindström, CarlAbstract Object detection is an important part of many autonomous driving systems, providing condensed information about the vehicle’s surroundings. For good performance in different environmental conditions, multi-modal object detection is often used, where information from different sensors are fused. Due to factors such as sensor noise, occlusion, and adverse weather conditions, there is an inherent uncertainty in the object detection task. Most state-of-the-art approaches for multi-modal 3D object detection do not model these uncertainties for regression. Explicitly modeling and estimating the uncertainties leads to higher interpretability, allows analysis of difficult situations, and can improve the performance of downstream tasks. In this work, we explore how uncertainty can be modeled and estimated in multimodal 3D object detection. We show that directly modeling the uncertainties of bounding box parameters can provide meaningful uncertainty estimates without sacrificing neither predictive performance nor computational efficiency. We compare modeling the uncertainties both separately per detection, using normally distributed random vectors, and jointly per frame, using Poisson multi-Bernoulli random finite sets. Our results show that separate modeling enhances predictive performance, while joint modeling yields more accurate uncertainty estimates. Additionally, we demonstrate that these predicted uncertainties can identify unlabeled data where the model performs poorly, underscoring their importance for more interpretable and safe autonomous driving systems.
- PostMIMO-waveforms for GMTI radar - A study on the use of MIMO waveforms to decrease MDV in GMTI radar(2024) Jisonsund, Max; Chalmers tekniska högskola / Institutionen för elektroteknik; McKelvey, Tomas; Dammert, PatrikAbstract This study investigates the use of MIMO waveforms in mitigating the effects of clutter in GMTI radar systems. Clutter, which is any unwanted reflection from the ground with a Doppler shift will affect the performance of a radar. In particular the minimum detectable velocity will increase. By applying MIMO waveforms it is possible to, in some cases, increase performance and decrease the minimum detectable velocity. This study has through simulations explored MIMO systems and compared these to traditional SIMO systems. Furthermore sparse arrays has also been explored and once again compared between MIMO and SIMO. Moreover this study has developed a unified model capable of simulating systems using for example SIMO, MIMO or sparse arrays. This serves as a valuable tool to perform further studies on the topic.
- PostModeling and Control of Inductive Power Transfer System with Misalignment and Load Independence(2024) Lai, Linhua; Chalmers tekniska högskola / Institutionen för elektroteknik; Liu, Yujing; Tang, JunfeiAbstract Inductive Power Transfer (IPT) presents a promising solution for electric vehicles (EVs), yet challenges persist in achieving high tolerance against coupling and load variations. To address this issue, new model and control scheme for IPT system are required. This thesis is dedicated to the modeling and control of an IPT system with a focus on misalignment and load tolerance. To understand the properties and performance of the IPT system, this thesis begins with an investigation into the conventional model of an IPT system with a seriesseries resonant tank. Using the Finite Element Method (FEM), the electromagnetic field around the magnetic coupler is studied, providing insights into the relationship between misalignment and mutual inductance. Subsequently, an equivalent circuit is constructed for the IPT system. It becomes evident that under the conventional control scheme, the system’s output power and efficiency are vulnerable to misalignment variations, posing a significant obstacle in power flow management. To address this challenge, a novel digital control strategy is proposed. By synchronizing the inverter output voltage with the transmitter current, constant output power can be achieved in high coupling conditions. The control circuit is simple, as it only captures the zero crossing of the current in the transmitting coil. However, inherent propagation delays in the measurement and control circuit can lead to frequency drift and high switching losses, thereby compromising system efficiency and output power. In response, a digital control scheme with delay compensation is introduced, and the boundary of soft switching operation is investigated. It is concluded that soft switching operation can be achieved by applying the new control strategy. Experimental results reveal that within a misalignment range of 0−160mm, the system maintains nearly constant output power. A peak efficiency of 92% can be attained when soft switching is realized in the inverter. This proposed control scheme enhances the robustness of the IPT system against misalignment variations, facilitating cost-effective hardware implementation and obviating the need for dualside communication. However, the variation of load during power transfer can still cause change in output power. Further analysis of power flow under various load conditions reveals the system’s sensitivity to load variations. Leveraging the grid-connected rectifier, the DC-link voltage, crucial for the IPT system’s output power, can be controlled to address the issue. Thus, the voltage controller of a three-phase Vienna rectifier is designed. Simulation results demonstrate that the rectifier has less than 1V output voltage ripple under 800V DC-link output voltage and 50kW output power, with a total harmonic distortion of input current of 2.67%. Notably, the stability of the IPT system remains unaffected by variations in the DC-link voltage. Furthermore, the proposed control strategy provides more flexibility to the charging system without requiring additional DC-DC converters. This thesis addresses the key challenges in IPT system design and control, offering innovative solutions to achieve robust, efficient, and adaptable EV charging systems. The proposed control schemes provide constant output power despite changes in coupling and load conditions. Therefore, the works in this thesis pave the way for the development of versatile, compact, and lightweight IPT systems.
- PostAutomated Simulation Environment for Enhanced Radar Technology Learning Creating Scenarios using Deep Machine Learning and Stochastic Target Trajectories for Future Radar Simulation(2024) Olsson, Louise; Wagné, Johanna; Chalmers tekniska högskola / Institutionen för elektroteknik; McKelvey, Tomas; Forsberg, ErikAbstract Simulations contribute to reduced resource requirements, faster testing of new equipment, and an often overlooked advantage: enhanced learning through interactive scenarios, immediate feedback, and practical experience in a safe environment. Whilethe conservation of resources contributes to sustainability, simulations also provide a quicker and more visual way to learn, which is particularly beneficial in complex fields such as radar technology. This accelerated learning process can expedite new employee onboarding, enabling more time to be dedicated to critical tasks where most needed. An automated simulation environment that generates new scenarios each use would further enhance these benefits. This study explores the development of such a system and how to implement it in the radar field. The work consisted of three main parts: creating a 3D environment based on real-world locations using deep machine learning, implementing automated target spawning within the 3D environment, and designing a simulation to demonstrate radar concepts for educational purposes. The creation of the environment requires a terrain map to predict height map and terrain type, achieved using semantic segmentation and a U-Net structure. The data from the network includes information about the terrain’s elevation and the corresponding surface color based on the terrain type. This data serves as the basis for creating a 3D environment using a Python 3D plotting library. Targets are created based on the environment, considering constraints such as maximum velocity, turn rate, and collision avoidance. Combining all the parts was a substantial aspect of the work. Before the system can function solely by inputting a terrain map, further improvements are necessary in the form of automatic data transfer and further dynamics implemented for the targets. Currently, the transfers between the main parts still need to be done manually. The goal is for the system to operate as an automated chain where one part feeds the next part with the necessary information. In summary, the idea of developing a fully automated simulation has been demonstrated to be effective through the tests conducted in this thesis. While the results of the project can support future research in the field, still many challenges await, such as ensuring that all parts of the simulation can automatically transfer data.
- PostSolar PV and Battery Systems In Sweden: An investigation of the frequency ancillary market for batteries(2024) Flory Kjellin, Karolina; Olofsson, Maja; Chalmers tekniska högskola / Institutionen för elektroteknik; Ehnberg, Jimmy; Etherden, NicholasAbstract The aim of this thesis was to evaluate the possibilities of tackling the constant changes in the market for frequency ancillary services from batteries, both from a short and long-term perspective. The short-term perspective focused on the possibilities of combining supplying FCR-D to the grid with increasing self-consumption of electricity from solar PV in private homes. This was done in two steps, first historical frequency disturbances were investigated to evaluate the possibilities of increasing the use of batteries thst provide FCR-D. The seond step was done by implementing an algorithm for battery operation that combined FCR-D with increasing self-consumption that was tested on real customer data. The long-term perspective instead focused on service development processes for companies in aggregator positions to give recommendation on how to continuously work with service development to be ready for sudden changes in the market. The result for the short-term perspective showed that 99.7% of disturbances are below five minutes in duration and batteries providing FCR-D never reach any critical levels of not being able to deliver FCR-D during a disturbance. Further, the results from simulating combinations of providing FCR-D and increasing self-consumption showed that a simultaneous operation yielded the highest self-consumption. Using 34% of inverter capacity which had been reserved for recharging gave the highest self-consumption result. Recommendations in the long-term perspective include maintaining close customer relationships and involving them in the service development process, as well as investing in market sensing and structured service development.