Our research group recently participated in the ACM International Conference on Neuromorphic Systems (ICONS 2026) in Chicago, presenting two projects in neuromorphic biomedical systems and embodied NeuroAI. 

Our paper, “Neuromorphic Silicon Neuron Controller for Adaptive Deep Brain Stimulation in Parkinson’s Disease,” was presented orally and selected as one of the Top 3 Best Paper Finalists. We introduced SiLIF-DBS, a neuromorphic silicon-neuron controller for adaptive deep brain stimulation in Parkinson’s disease. Closed-loop and adaptive simulations showed effective suppression of pathological beta activity, demonstrating the potential of low-power neuromorphic systems for adaptive DBS. 

We also presented the poster “Neuromorphic Embodied AI Robot for Place Cell Modeling with Rodent Experiment Calibration,” by Tianze Liu, Zihan Jin, Hongyu An, and Dong Song. This project explores how biologically inspired neural ensembles enable an neuromorphic robot to develop place-cell-like spatial representations through interaction with its environment. The model combines visual cues with spatial signals through associative learning and is compared with hippocampal recordings from neural signals of 4 rodents. 

It is my great pleasure to attend ISQED this week. 

It was great to catch up with old friends and meet new people working on circuits, AI, and neuromorphic systems. 

I also had the opportunity to visit our collaborator, Dr. Corina’s lab at UC Davis, where I learned about their fascinating ideas and contributions in adaptive deep brain stimulation. These in-person interactions were highly valuable and further strengthened our enthusiasm for future collaborations. I also had the opportunity to meet Prof. Deliang Fan from Arizona State University and Prof. Haitong Li from Purdue University and to learn about their recent work, which provided additional perspectives and inspiration for my own research and potential collaborations. 

Congratulations to Tianze Liu on receiving the Matt Wolfe Outstanding Graduate Research Assistant Award. Tianze Liu has made substantial contributions to the neuromorphic robotics research. He has served as the first author on multiple peer-reviewed publications, including one in Frontiers in Neuroscience (h5-index 118), a widely recognized journal in the neuromorphic community. 

Matt Wolfe was a 1992 BSEE graduate and MS degree candidate in Electrical Engineering. His family established this endowed award in his memory and is awarded each year to the ECE Department's outstanding research assistant.

It has been my great pleasure to work with an outstanding team at the Air Force Research Laboratory this summer. I am deeply grateful for the remarkable contributions of our interns. Your fresh insights and dedication have made a meaningful impact on our work. The participating students included Corey Decker from MIT, Ava McCarthy from the University at Albany, Musong Kwon from Texas A&M University, and Brody England from Northern Arizona University. 

After two years of effort, we successfully demonstrated associative learning on the neuromorphic robot at AFRL. With this capability, the robot can perform not only unsupervised learning but also transfer learning. All of these capabilities have been validated through on-field experiments.

We are excited to announce that, for the first time, our research on Neuromorphic Embodied AI system and Associative Learning was presented at the 2024 Society for Neuroscience (SfN) Annual Meeting in Chicago, Illinois. 

The poster, titled "Embodied Neuromorphic Robot System for Replicating Associative Learning of Rodents in an Open-Field Maze," was authored by Tianze Li, Yan Zhang, and Hongyu An.

This study presents an embodied neuromorphic system as an alternative to animal models for investigating associative learning. By incorporating computational models of spatial cells, such as place and grid cells, into a mobile robotic platform, the system was validated through both simulations and experiments. The results show that the embodied neuromorphic system can: (1) generate the characteristic firing patterns of place and grid cells while navigating a circular open-field arena; (2) replicate rodent associative learning by correlating visual cues with adverse stimulus locations; and (3) produce a learning curve that closely aligns with experimental data from animal models.

Time flies! After almost five years, our group attended the International Conference on Neuromorphic Systems (ICONS) 2024 in Washington, D.C. It was great to reconnect with old friends and meet new ones. We presented our latest work on neuromorphic closed-loop Deep Brain Stimulation systems and neuromorphic embodied AI projects: 

• Ananna Biswas, Hongyu An, "Preliminary Results of Neuromorphic Controller Design and a Parkinson's Disease Dataset Building for Closed-Loop Deep Brain Stimulation" arXiv preprint arXiv:2407.17756 (2024). 

• Parkinson's Electrophysiological Signal Dataset (PESD). 

• Mimicking Associative Learning of Rats via a Neuromorphic Robot in Open Field Maze using Spatial Cell Models. Tianze Liu, Md Abu Bakr Siddique and Hongyu An. in publishing. 

It is my great pleasure to have a wonderful team at the Air Force Research Lab this summer. I am deeply grateful for the remarkable efforts of our interns, including the talented Young Hokies from Virginia Tech. Your fresh insights and commitment have made a significant impact on our work.

A heartfelt thanks to the exceptional leadership of Dr. Bai. Your mentorship has been instrumental in our progress.

Additionally, it is my pleasure to provide a short course on neuromorphic robotic systems, helping to spread this cutting-edge research to a broader community.

Congratulations to Ananna Biswas for receiving NSF AccelNet NeuroPAC fellowship Award.  NeuroPAC is an NSF-supported effort under the AccelNet (Accelerating Research through International Network-to-Network Collaborations) program, led by the University of Maryland and Johns Hopkins University. Its goal is to network internationally neuromorphic engineers with computational neuroscientists, roboticists, and researchers of computational theory, control, and perception to advance the foundations of Neuromorphic Intelligence. 

We deeply appreciate the support from NSF and the NeuroPAC committee. We hope Ananna will have a fruitful research experience at the University of Southern California.

Congratulations to Md Abu Bakr Siddique for successfully defending his master's thesis titled "The Integration of neuromorphic computing in autonomous robotic systems" on April 5, 2024. His thesis research implements associative learning with an autonomous robot for an online learning scenario. This system emulates the classic associative learning of rats in T-maze. Abu Bakr will join the University of Florida for his doctoral study. We wish Abu good luck in his future research. 

Really appreciate the hard work of the students throughout 2022 and 2023!  We enjoyed the time and dinner with Dr. Zhang's group and look forward to more collaborations and a more fruitful 2024!

Congratulations to Noah Zins on receiving the Matt Wolfe Outstanding Graduate Research Assistant Award! 

Matt Wolfe was a 1992 BSEE graduate and MS degree candidate in Electrical Engineering who was tragically killed in an automobile accident during his second year of graduate study. He was an outstanding student; his research work and interactions with other students, both graduate and undergraduate, were exemplary. His family established this endowed award in his memory and is awarded each year to the ECE Department's outstanding research assistant.

Congratulations to Noah Zins for successfully defending his master's thesis titled "Neuromorphic Computing Applications in Robotics" on Aug. 2, May 2023. His thesis research implements associative learning with an Unmanned Ground Vehicle (UGV) and neuromorphic hardware (Intel’s Loihi) for an online learning scenario. This system emulates the classic associative learning in rats using the UGV in place of the rats. Specifically, it successfully reproduces fear conditioning with no pretraining procedure or labeled datasets. The UGV is rendered capable of autonomously learning the cause-and-effect relationship of the light stimulus and vibration stimulus and exhibiting a movement response to demonstrate memorization. Hebbian learning dynamics are used to update the synaptic weights during the associative learning process. The Intel Loihi chip is integrated with this online learning system for processing visual signals with a specialized neural assembly. Noah will join the industry. We wish Noah good luck in his industry career!

After three years of the COVID-19 pandemic, we are immensely pleased that we attended the ISQED in San Francisco, CA, in April 2023. We presented our latest work, "Reproducing Fear Conditioning of Rats with Unmanned Ground Vehicles and Neuromorphic Systems," which demonstrates the fear conditioning of rats using neuromorphic computing and mobile robots.

Congratulations to Noah Zins on receiving the Outstanding Graduate Student Teaching Award!