Acta Electronica Sinica, 2025
This paper proposes a bio‑plausible neural network that integrates motion and disparity pathways to achieve robust looming perception and real‑time collision detection, significantly improving depth feature extraction and adaptability in complex environments.
Neurocomputing, 2025
Inspired by insect directionally selective neurons, this study proposes and optimizes a binocular neuromorphic vision model that integrates depth-disparity processing to enhance looming detection and collision prediction.
Biomimetics, 2024
This work modeled the optimized LGMD model with contrast-polarity-specific inhibition to improve collision detection efficiency and tested it in embedded robotics.
The 14th International Conference on Biomimetic and Biohybrid Systems (Living Machines 2025), 2025
This work focus on modeling self-inhibition in neural networks for collision perception against translating motion.
International Joint Conference on Neural Networks (IJCNN), 2025
This work models the lobula plate/lobula columnar, type 2 (LPLC2) neural ensemble of fruit fly Drosophila, enabling robust multi-object looming detection and localization via a multi-attention mechanism.