BROWSE
Detail
Conference Paper
안내견 로봇 내비게이션 기술 검증을 위한 시뮬레이터 구현
Cited - time in 
Share 
Share
- Issue Date
- 2026-07
- Citation
- 제어·로봇·시스템학회 학술 대회 (ICROS) 2026, pp.901-902
- Publisher
- 제어·로봇·시스템학회
- Language
- Korean
- Type
- Conference Paper
- Abstract
- Developing quadrupedal guide dog robots in real-world environments involves high costs and safety risks. This work presents a high-fidelity navigation research platform using NVIDIA Isaac Sim and ROS2. Packaged as a dual-container Docker system, it integrates a quadruped robot model, simulated sensors (RGB-D, LiDAR), and GPU-accelerated traversability mapping. Overcoming the visual and sensory limitations of conventional simulators, the proposed platform achieves both visual and dynamical realism. Results validate its effectiveness for testing navigation algorithms, providing an accessible and robust open-source environment for quadrupedal robots.
- Keyword
- robot simulator, robot navigation, sim-to-real
- KSP Keywords
- GPU-accelerated, High-fidelity, Open source, Quadruped Robot, RGB-D, Real-world, Research Platform, Robot Navigation, Robot simulator, navigation algorithm