A5002826974- Autonomous Vehicle Technology and Safety
- Advanced Memory and Neural Computing
- Reinforcement Learning in Robotics
- Ferroelectric and Negative Capacitance Devices
- Anomaly Detection Techniques and Applications
- Traffic Prediction and Management Techniques
- Neural dynamics and brain function
- Hate Speech and Cyberbullying Detection
- Neural Networks and Applications
- Fault Detection and Control Systems
- Traffic control and management
University of Colorado Boulder
2023
Oak Ridge National Laboratory
2020-2021
Neuromorphic computing has many opportunities in future autonomous systems, especially those that will operate at the edge. However, there are relatively few demonstrations of neuromorphic implementations on real-world applications, partly because lack availability hardware and software, but also an accessible demonstration platform. In this work, we propose utilizing F1Tenth platform as evaluation task for computing. is a competition wherein one tenth scale cars compete racing task;...
There have been many recent advancements in imitation and reinforcement learning for autonomous driving, but existing metrics generally lack the means to capture a wide range of driving behaviors compare severity different failure cases. To address this shortcoming, we introduce Quan-titative Evaluation Driving (QED), which assesses aspects behavior including ability stay center lane, avoid weaving erratic behavior, follow speed limit, collisions. We scores generated by QED against assigned...
Identifying expressions of human values in textual data is a crucial albeit complicated challenge, not least because ethics are highly variable, often implicit, and transcend circumstance. Opinions, arguments, the like generally founded upon more than one guiding principle, which necessarily independent. As such, little known about how to classify predict moral undertones natural language sequences. Here, we describe present solution ValueEval, our shared contribution SemEval 2023 Task 4....
We repurposed an adversarial evolutionary algorithm, Gremlin, from finding driving scenarios where a model of autonomous vehicle drove poorly to troubleshooting quality evaluation criteria. evaluated the performance "perfect driver" robot in virtual town environment using same fitness criteria intended for deep learner (DL) trained driver. found that handled turns, and used Gremlin iteratively improve were confident could then be applied DL-based models as originally intended, this approach...