Chalmers Open Digital Repository

Välkommen till Chalmers öppna digitala arkiv!

Här hittar du:

Studentarbeten utgivna på lärosätet, såväl kandidatarbeten som examensarbeten på grund- och masternivå
Digitala specialsamlingar, som t ex Chalmers modellkammare
Utvalda projektrapporter

Forskningspublikationer, rapporter och avhandlingar hittar du i research.chalmers.se

Enheter i Chalmers ODR

Välj en enhet för att se alla samlingar.

Visar 1 - 2 av 2

Senast publicerade

Integrating a Neuro-Symbolic Planner for Safe Driving
(2026) Alvandi, Saghar; Wingård, Jennifer
Ensuring the safety and reliability of autonomous driving systems remains an ongoing challenge. The complexity of real-world traffic scenarios highlights the importance of autonomous driving systems being able to reason about their surrounding environments and make safe decisions. This thesis proposes a modular neuro-symbolic framework that decouples trajectory prediction from motion planning. Rather than relying solely on an end-to-end deep-learning-based planner, the proposed approach uses a fine-tuned large language model to predict the future trajectories of surrounding actors. These predictions are incorporated as dynamic obstacles into a symbolic A* planner, which generates a goal-oriented trajectory for the ego vehicle subject to safety constraints. The proposed framework is evaluated on data generated from driving scenarios developed in Scenic and executed in the CARLA simulator, with its performance compared against an end-to-end deep-learning-based planner using the same set of 144 scenarios. The results demonstrate that this modular neuro-symbolic approach achieves a higher rate of safe, goal-reaching trajectories, with a success rate of 59.0% compared with 27.1% for the baseline planner. However, the neuro-symbolic planner also exhibits a higher collision rate of 32.6%, compared with 10.4% for the baseline planner. This difference is partly due to the baseline frequently producing safe but non-goal-reaching trajectories, with 57 scenarios falling into this category because the ego vehicle remains at least 100 meters away from the other actors throughout the scenario. These findings highlight a trade-off between collision avoidance and goal achievement, indicating the need for further evaluation to draw clearer conclusions regarding the safety of the approach.
Structure-borne noise assessment of automotive componentsusing theblocked force approach
(2026) Man, Nicholas
Changes in the vehicle industry have heightened the demand for performing acoustic and vibration analysis earlier in the product development process. This is challenging as NVH issues are often identified during full system integration, when design changes are costly and limited due to frozen vehicle configurations. To address this, virtual prototypes can be developed where components are independently characterized at their interfaces and transferred between assemblies. A major limitation of this approach is that vibration forces from source components are not inherently independent, as interface forces obtained through conventional methods depend on boundary conditions and are therefore not transferable between systems. To over come this, the in-situ blocked force method has been introduced to characterize source components such that the resulting interface forces are intrinsic to the source and independent of the receiving structure. This thesis presents a case study applying the in-situ blocked force method to an electro-hydraulic steering pump as a source component. The objectives are to validate the in-situ blocked force method, demonstrate that the derived blocked forces are independent of the receiving structure and transferable between systems, and establish a workflow for digital twin based acoustic and vibration analysis. The results show good consistency of the derived blocked forces across two dynamically di!erent receiving structures. This consistency is confirmed through cross-validation between test configurations. Furthermore, the developed digital twin enables component-based transfer path analysis and the simulation of modification scenarios
Semantic Communication for Closed-loop Sensing-Communication-Control Systems
(2026) Wang, Jingcheng
This thesis investigates channel state information (CSI) as a sensing input for position estimation and simulated closed-loop control, motivated by the Level-2 interpretation of semantic communication. It evaluates task-relevant state extraction rather than a modeled semantic transmission link. A simulation pipeline combines Sionna RT channel generation in a street-canyon scene, a ResNet-based estimator, Kalman filtering, and linear quadratic regulator (LQR) control. Complex CSI is represented by real and imaginary components across antennas and subcarriers. The neural model predicts two-dimensional position and coordinate-wise variance. Position predictions supply measurements to the filter, while the controller acts on the filtered state. The learned covariance is evaluated separately; the main closed-loop experiments use fixed diagonal covariance. On a random held-out test split of 2000 samples within the simulated scene, the estimator achieves a mean Euclidean localization error of 0.857 m, an RMSE of 1.204 m, and a median error of 0.673 m. Joint antenna–subcarrier masking yields an accuracy plateau: retaining 56.25% of the full CSI coefficients gives the lowest observed mean error, 0.626 m. This does not establish bitrate or energy savings. Predicted covariance is weakly correlated with localization errors and fails the reported consistency check, limiting its direct use in Kalman filtering. The fixed-covariance sweep demonstrates tracking sensitivity to measurement confidence. In one paired seed-0 rollout per setting, a variance of 1 m2 per coordinate yields the lowest true-state LQR cost, 750.27, and tracking RMSE, 0.683 m, among the tested settings. These findings characterize within-scene localization, uncertainty limitations, and covariance-sensitive tracking. Ideal interfaces, one scene and trajectory, and missing controller controls restrict conclusions about communication benefits, generalization, and the contribution of individual pipeline components. Calibrated uncertainty and broader evaluation are needed before deployment.
Multi-View Pedestrian Detection Without Scene-Specific Fine-Tuning via 3D Foundation Models
(2026) Lai, Bowen; Zhang, Junchao
Multi-view pedestrian detection in occluded scenes often relies on scene-specific 3D reconstructions, which are accurate but require lengthy optimization. This thesis investigates whether an off-the-shelf 3D vision foundation model (MapAnything) can provide metric 3D point clouds for this task without scene-specific fine-tuning. We first characterize the foundation model’s geometric error, finding that calibrated inputs yield manageable scale drift correctable via ground-plane scaling, whereas uncalibrated inputs cause severe frame-to-frame distortion (averaging 66.0 cm spatial drift), degrading detection accuracy from 59.9 MODA to 29.1 MODA. To integrate this foundation-model geometry into a detector, we build a differentiable bird’s-eye view (BEV) pipeline based on Lift-Splat-Shoot and PointPillars. To handle the 30–80 cm point cloud dispersion typical of foundation models, we use dilated convolutions to enlarge the BEV backbone’s receptive field. This architecture achieves 68.7 MODA on Wildtrack and 64.0 MODA on MultiviewX. Compared to a per-scene optimized baseline (DCHM), which reaches 83.0 MODA in our singleiteration supervised reproduction and approximately 95 MODA in the best supervised results reported by its authors, the gap reflects the limitations of off-the-shelf foundation geometry. Finally, we show that auxiliary mechanisms like soft-depth sampling can cause oversmoothing, and cross-subset experiments reveal that BEV convolutional networks severely overfit to the training camera rig layout (dropping up to 28.35 MODA when testing on unseen camera subsets).
Interactive Simulation Systems for Industrial Equipment - Supporting System Verification Through Interactive Information Visualization
(2026) Persson, Jakob; Tan Quach, Kevin
Industrial process systems rely on precise level measurements to run safely and profitably. To verify that these measuring instruments work correctly under realistic conditions, engineers use simulation setups like the Interactive Plant Environment (IPE). However, exploring the massive amounts of live data generated during these tests is difficult without specialized tools. The primary purpose of this project is to answer how interactive simulation systems should be designed to support system verification. To investigate this, a prototype interface was developed to allow test engineers to easily run automated scripts, control the system manually, and analyze real-time data. Using a Research through Design (RtD) approach and the Double Diamond process, the project started with semi-structured interviews to understand the daily work flows and frustrations of expert test engineers. Insights from this pre-study indicated that standard usability checklists are insufficient for live industrial environments. By combining these empirical findings with foundational Human-Computer Interaction (HCI) principles and Information Visualization (InfoVis) theory, the project introduces a set of custom design guidelines: spatial fidelity and dynamic state mapping, flexible test control, stable views for live data, clear error handling, proactive input validation and defensive safety nets, and data filtering. Based on these guidelines, a functional web application prototype was built featuring visual tank models, live data plotting, and seamless switching between automated and manual execution. A heuristic evaluation of the prototype suggested that the interface supports expert test engineers by reducing cognitive load through visual metaphors and maintaining stable views during live data inspection. While the prototype managed complex testing scenarios, it also exposed vulnerabilities regarding conflicting commands and unguided error recovery, which directly resulted in refining and expanding the guidelines. Ultimately, this project proposes a set of domain-specific guidelines that can assist designers and developers in building interactive simulation and verification systems for similar industrial contexts.