Max Robotics — Global Service Robot Directory + US Certification

About

Jingkai Sun is an emerging robotics and artificial intelligence researcher whose work sits at the intersection of humanoid robot control, reinforcement learning, and trajectory prediction. His most impactful contributions focus on enabling humanoid robots to achieve natural, human-like locomotion through innovative learning paradigms. His 2024 paper "Whole-body Humanoid Robot Locomotion with Human Reference" (31 citations) addresses the fundamental challenge of designing reward functions for full-body humanoid control by leveraging human motion as a guiding reference. Complementing this, his work on LLM-based humanoid control through quantized imitation learning (28 citations) demonstrates a forward-thinking approach to generalizing robot behavior across diverse tasks using large language models. Beyond humanoid robotics, Sun has made notable contributions to pedestrian trajectory prediction, developing spatial-temporal models using gated linear units (21 citations) and dynamic target-driven networks for anticipating pedestrian movement in autonomous driving and navigation contexts. His 2025 work on fully spiking neural networks for legged robots signals a pioneering interest in neuromorphic computing for robotics applications. Collectively, Sun's research reflects a rare breadth spanning embodied AI, deep reinforcement learning, and predictive modeling, positioning him as a versatile contributor to next-generation intelligent robotic systems.