본문 바로가기
  • Home

A Study on the Basic Control of Humanoid Robot

  • Journal of Software Assessment and Valuation
  • Abbr : JSAV
  • 2025, 21(2), pp.89~101
  • Publisher : Korea Software Assessment and Valuation Society
  • Research Area : Engineering > Computer Science
  • Received : June 1, 2025
  • Accepted : June 20, 2025
  • Published : June 30, 2025

Dong-Ha Hwang 1 Kim Jae Woong 2 Kim Dong Hyun 3

1공주대학교
2국립공주대학교
3나사렛대학교

Accredited

ABSTRACT

This paper presents the implementation and experimental analysis of a fundamental control system for humanoid robots utilizing Python-based SDK and API frameworks. As humanoid robots gain increasing adoption across diverse applications due to their anthropomorphic design and motion capabilities, the development of precise and modularized control technologies becomes essential. This study implements and validates core functionalities including locomotion, arm manipulation, and balance maintenance, with comprehensive evaluation of control accuracy and system responsiveness. A real-time sensor data acquisition and processing system was developed through ROS integration, specifically targeting LiDAR and IMU sensor streams. The sensor integration experiments were systematically evaluated based on three key performance metrics: latency, data stability, and processing efficiency. Furthermore, the interaction between sensor inputs and control commands was modularized to establish a framework compatible with reinforcement learning-based autonomous control algorithms. Experimental results demonstrate the system's effectiveness in achieving reliable control performance with average response times of 0.2 seconds for locomotion commands and positioning accuracy within 5% error margins. The developed framework provides a robust foundation for advanced autonomous control systems and multi-modal sensor fusion applications in humanoid robotics.

Citation status

* References for papers published after 2023 are currently being built.