Publications
Chosen filters:
Chosen filters:
Making model checking feasible for GOAL KU Leuven
Safe and Robust Robot Behavior Planning via Constraint Programming KU Leuven
The safe operation of an autonomous system is a complex endeavor, one pivotal element being its decision-making. Decision-making logic can formally be analyzed using model checking or other formal verification approaches. Yet, the non-deterministic nature of realistic environments makes these approaches rather troublesome and often impractical. Constraint-based planning approaches such as Tumato have been shown to be capable of generating ...
Effectiveness of Forward Error Corrections Over Different Wired Communication Channels in Harsh Electromagnetic Environments KU Leuven
This paper presents the effectiveness of several forward error corrections in a harsh electromagnetic environment. A harsh electromagnetic environment consists of many reflections and unknowns which could negatively impact the functioning of the operational equipment or systems therein. Accordingly, four printed circuit boards with varying EMC designs are considered to emulate four single-trace communication channels. In addition, based upon the ...
Resilience of Reed-Solomon Codes Against Single-Frequency Electromagnetic Disturbances: Fault Elimination Through Encoder Tuning KU Leuven
In increasingly electromagnetic-polluted environments, communication networks are becoming more vulnerable. Even networks equipped with error control techniques suffer from this problem. Electromagnetic disturbances can result in corrupted data which are undetectable by error control techniques. Such scenarios are extremely dangerous as the system is unaware of the corruption. This could lead to critical failures. Thus, protecting communication ...
Towards classification trustworthiness: one-class classifier ensemble KU Leuven
Many autonomous safety-critical systems rely on neural networks for image classification. While they achieve high accuracy, their decisions are hard to interpret. Also, a known issue with neural networks is that they tend to provide high probabilities for unknown images. Uncertainty on how neural networks will behave is a challenge to safety. To address these issues, this paper presents a different approach by using an ensemble of one-class ...