JAESIK JEONG

This paper presents heuristic design methodologies for humanoid robots developed through practical experience in international humanoid robot competitions. Since humanoid robot development requires close integration between design and engineering, this study proposes an approach that connects technical requirements with design-oriented decision-making. The proposed approach focuses on mechanical and electrical design strategies derived from repeated prototyping, testing, and competition-based refinement. For mechanical design, this paper introduces lightweight structural concepts and optimized wiring layouts using hollow shafts. These strategies improve robustness, reduce unnecessary weight, and support efficient dynamic movement in diverse environments. For electrical design, the paper explains component selection based on performance, reliability, and operational efficiency. Main and sub controllers, sensors, actuators, and communication modules are selected to support control algorithms, artificial intelligence (AI), real-time data processing, and stable communication. Based on practical experience, this paper discusses key challenges in humanoid robot development and practical solutions obtained through iterative improvement. By sharing design strategies grounded in real-world implementation, this research offers useful insights for developing effective humanoid robot platforms.

Jee-Hyun Yang, JAESIK JEONG Verified by AVIS CertificateFIRA World Summit 2026Submitted 5 Jun 2026

This study investigates deep reinforcement learning for bipedal robot gait control using NVIDIA Isaac Gym. A high-degree-of-freedom simulation is developed, and the agent first learns under an unconstrained baseline by directly controlling joint angles. To improve stability and realism, inverse kinematics and Zero Moment Point constraints are progressively introduced, ensuring feasible motion and center of mass balance within the support polygon. Some experiments further employ reference foot trajectories to guide learning and enhance efficiency. The effects of different control conditions—guided versus unguided and with or without constraints—are analyzed to understand how knowledge-based limitations influence the agent’s learning performance and gait behavior.

Chih-Cheng Liu 劉智誠, JAESIK JEONG, CHENG LIN KUO Verified by AVIS CertificateFIRA World Summit 2026Submitted 4 Jun 2026

This paper applies deep reinforcement learning to unmanned vehicles in a simulated environment, aiming to overcome the limitations of traditional path planning methods. The work is divided into two parts: (1) Q-Learning with known environmental information, and (2) Deep Q-Learning for unknown environments. In the Q-Learning approach, the agent’s coordinates in the simulation are used as states to build a Q-table and find the optimal path. For Deep Q-Learning, a neural network is built with PyTorch, and Experience Replay and Fixed Q-targets are used to improve training stability. Experiments verify that, using the weights trained in simulation, the unmanned vehicle can automatically avoid obstacles and reach the destination in a real-world environment.

TSAI CHENG EN, Chih-Cheng Liu 劉智誠, JAESIK JEONGFIRA World Summit 2026Submitted 4 Jun 2026