A Bio-Plausible Visual Neural Network for Locust-Inspired Collision Perception
Abstract
Locust visual systems have long served as an important biological paradigm for studying looming perception and collision avoidance. Numerous computational models have successfully reproduced the selective responses of Lobula Giant Movement Detector (LGMD) neurons to approaching objects, thereby emulating the fundamental functionality of the biological system. However, existing models remain limited in biological plausibility and robustness when operating in complex and dynamic visual environments. To address these limitations, we propose a biologically plausible neural network for locust-inspired looming detection. The proposed framework incorporates a spatially isotropic sampling strategy that mimics the ommatidial organization of the locust compound eye, a population-voting mechanism inspired by population coding in biological neural systems, and leaky integrate-and-fire neuronal dynamics to replace conventional sigmoid-based membrane activation. Systematic experiments on synthetic stimuli, laboratory sequences, and real-world driving scenarios demonstrate that the proposed model improves robustness under challenging visual conditions while preserving computational efficiency and enhancing biological fidelity. These results highlight the potential of biologically grounded neural computation for robust and efficient collision perception.
Related
- Seeing Through Uncertainty: Free-Energy-Inspired Real-Time Adaptation for Robust Visual Navigation
- High-dimensional population codes reveal interpretable and diverse features underlying visual perception
- Attention selectively enhances collision sensitivity in the superficial SC, PBGN and centromedial amygdala to looming stimuli in the upper visual field
- Seeing danger? Visual constraints and collision risk in the Bearded Vulture Gypaetus barbatus
- A visual perception-guided framework for camera path planning in large-scale digital twin and AI-generated 3D scenes
BibTeX
@misc{fu2026bioplausible,
title = {{A Bio-Plausible Visual Neural Network for Locust-Inspired Collision Perception}},
author = {Qinbing Fu and Jiani Li and Jiajun Huang and Jigen Peng},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.10183},
eprint = {2609.10183},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.10183},
note = {CC BY-NC-ND 4.0},
}
RIS
TY - UNPB
TI - A Bio-Plausible Visual Neural Network for Locust-Inspired Collision Perception
AU - Qinbing Fu
AU - Jiani Li
AU - Jiajun Huang
AU - Jigen Peng
PY - 2026
DA - 2026/09/09
PB - arXiv
DO - 10.48550/arxiv.2609.10183
UR - https://arxiv.org/abs/2609.10183
AB - Locust visual systems have long served as an important biological paradigm for studying looming perception and collision avoidance. Numerous computational models have successfully reproduced the selective responses of Lobula Giant Movement Detector (LGMD) neurons to approaching objects, thereby emulating the fundamental functionality of the biological system. However, existing models remain limited in biological plausibility and robustness when operating in complex and dynamic visual environments. To address these limitations, we propose a biologically plausible neural network for locust-inspired looming detection. The proposed framework incorporates a spatially isotropic sampling strategy that mimics the ommatidial organization of the locust compound eye, a population-voting mechanism inspired by population coding in biological neural systems, and leaky integrate-and-fire neuronal dynamics to replace conventional sigmoid-based membrane activation. Systematic experiments on synthetic stimuli, laboratory sequences, and real-world driving scenarios demonstrate that the proposed model improves robustness under challenging visual conditions while preserving computational efficiency and enhancing biological fidelity. These results highlight the potential of biologically grounded neural computation for robust and efficient collision perception.
ER -