Papers
9
Total Citations
142
H-Index
5
About
Taiping Zeng is a robotics and computational neuroscience researcher whose work sits at the compelling intersection of brain-inspired computing and autonomous robot navigation. His research focuses primarily on neurobiologically inspired simultaneous localization and mapping (SLAM), cognitive mapping, and visual odometry, drawing on discoveries from the entorhinal-hippocampal system — including grid cells, head-direction cells, and place cells — to design more robust and efficient navigation algorithms for mobile robots. Zeng's most influential contribution, "NeuroBayesSLAM" (2020, 52 citations), demonstrates how Bayesian integration of multisensory information, modeled on neural circuits, can significantly improve robotic localization in large-scale dynamic environments. His 2017 work on conjunctive representations of space and movement (34 citations) further established continuous attractor networks as a powerful framework for cognitive mapping. Alongside these theoretical advances, he has tackled practical challenges in autonomous exploration and visual self-localization, including a rapidly-exploring random trees approach (21 citations) and a spatiotemporal dual-stream visual odometry network. His more recent work on topometric mapping and Hidden Markov Model-based localization signals a broadening research vision toward richer, context-aware scene understanding for robots operating without GPS — a frontier challenge in embodied AI.
Research Focus
Key Achievements
Top Papers
- 1
- 2Cognitive Mapping Based on Conjunctive Representations of Space and Movement34 citations · 2017
- 3A brain-inspired compact cognitive mapping system23 citations · 2020
- 4
- 5Spatiotemporal Dual-Stream Network for Visual Odometry5 citations · 2025
- 6
- 7
- 8
- 9