Mai Nishimura
Papers
9
Total Citations
80
H-Index
4
About
Mai Nishimura’s research lies at the intersection of multi-agent systems, swarm robotics, and crowd-aware robot navigation, with a focus on enabling robots to operate safely and efficiently in dynamic, human-populated environments. Her most impactful work, “Prioritized Safe Interval Path Planning for Multi-Agent Pathfinding With Continuous Time on 2D Roadmaps” (29 citations), tackles the challenging problem of coordinating hundreds of agents on continuous-time roadmaps, offering a scalable solution that outperforms traditional discrete-time approaches. In a strikingly different direction, Nishimura introduced the concept of “embodied swarm robots” in her 2024 paper “Swarm Body” (19 citations), exploring how the human brain’s plasticity can integrate a collective of robots as an artificial body part for intuitive environmental interaction. She has also made significant contributions to crowd density forecasting (14 citations) and view birdification—recovering ground-plane crowd trajectories from ego-centric video—which is critical for mobile robot localization and navigation. Her work “L2B: Learning to Balance the Safety-Efficiency Trade-off in Interactive Crowd-aware Robot Navigation” (3 citations) further demonstrates her commitment to developing deep reinforcement learning frameworks that allow robots to navigate crowded spaces while balancing collision avoidance with task completion. Nishimura’s research consistently bridges theoretical advances with real-world robotic applications.
Research Focus
Key Achievements
Top Papers
- 1
- 2Swarm Body: Embodied Swarm Robots19 citations · 2024
- 3Crowd Density Forecasting by Modeling Patch-Based Dynamics14 citations · 2020
- 4
- 5
- 6
- 7Crowd Density Forecasting by Modeling Patch-based Dynamics3 citations · 2019
- 8
- 9FluidicSwarm: Embodiment of Swarm Robots Using Fluid Behavior Imitation1 citations · 2025