A5074876781- Advanced Memory and Neural Computing
- Robotics and Sensor-Based Localization
- Neural dynamics and brain function
- Advanced Neural Network Applications
- Wireless Signal Modulation Classification
- Autonomous Vehicle Technology and Safety
- Neural Networks and Reservoir Computing
- Real-time simulation and control systems
- CCD and CMOS Imaging Sensors
- Cryptographic Implementations and Security
- Electric and Hybrid Vehicle Technologies
- Adversarial Robustness in Machine Learning
- Digital Media Forensic Detection
- Vehicle Dynamics and Control Systems
Purdue University West Lafayette
2024
Vision-based object tracking is a critical component for achieving autonomous aerial navigation, particularly obstacle avoidance. Neuromorphic Dynamic Vision Sensors (DVS) or event cameras, inspired by biological vision, offer promising alternative to conventional frame-based cameras. These cameras can detect changes in intensity asynchronously, even challenging lighting conditions, with high dynamic range and resistance motion blur. Spiking neural networks (SNNs) are increasingly used...
Object detection and tracking is an essential perception task for enabling fully autonomous navigation in robotic systems. Edge robot systems such as small drones need to execute complex maneuvers at high-speeds with limited resources, which places strict constraints on the underlying algorithms hardware. Traditionally, frame-based cameras are used vision-based due their rich spatial information simplified synchronous sensing capabilities. However, obtaining detailed across frames incurs...
Vision-based object tracking is an essential precursor to performing autonomous aerial navigation in order avoid obstacles. Biologically inspired neuromorphic event cameras are emerging as a powerful alternative frame-based cameras, due their ability asynchronously detect varying intensities (even poor lighting conditions), high dynamic range, and robustness motion blur. Spiking neural networks (SNNs) have gained traction for processing events energy-efficient manner. On the other hand,...
Deep Neural Networks (DNNs) which are trained end-to-end have been successfully applied to solve complex problems that we not able in past decades. Autonomous driving is one of the most yet be completely solved and autonomous racing adds more complexity exciting challenges this problem. Towards challenge applying learning racing, paper shows results on two aspects: (1) Analyzing relationship between data used for training maximum speed at DNN can predicting steering angle, (2) network...
Vision-based object tracking is an essential precursor to performing autonomous aerial navigation in order avoid obstacles. Biologically inspired neuromorphic event cameras are emerging as a powerful alternative frame-based cameras, due their ability asynchronously detect varying intensities (even poor lighting conditions), high dynamic range, and robustness motion blur. Spiking neural networks (SNNs) have gained traction for processing events energy-efficient manner. On the other hand,...