Nannan Du

Papers

Total Citations

H-Index

About

Nannan Du is a researcher advancing the frontier of intelligent robotics, with a focus on deep reinforcement learning, motion planning, and autonomous assembly. Her work addresses critical challenges in enabling robots to navigate complex, dynamic environments and learn from human demonstrations. In her highly cited 2021 study, "Toward Obstacle Avoidance for Mobile Robots Using Deep Reinforcement Learning Algorithm," she enhanced the deep deterministic policy gradient (DDPG) algorithm by refining its experience replay mechanism, achieving superior continuous control for obstacle avoidance—a contribution that has garnered 4 citations. That same year, her paper "Learning from Demonstration Using Improved Dynamic Movement Primitives" tackled the problem of forcing term invalidation in motion learning, proposing a robust method for robots to acquire and adapt complex motion sequences, also earning 4 citations. Most recently, in 2025, she introduced "An Improved Searching Strategy Based on Contact State Recognition for the Assembly of Avionics Connector," a novel approach that leverages contact state recognition to boost precision in high-stakes assembly tasks. Du’s work is pivotal for developing more adaptive, reliable robots in manufacturing and service applications, blending algorithmic innovation with practical deployment.