Peixin Chang

University of Illinois Urbana-Champaign

Papers

7

Total Citations

226

H-Index

4

About

Peixin Chang is a robotics researcher whose work sits at the intersection of autonomous navigation, human-robot interaction, and multimodal learning. His research focuses primarily on enabling mobile robots to navigate safely and intelligently in complex, human-populated environments — a challenge that demands both perceptual sophistication and real-time decision-making. Chang's most influential contribution is his work on decentralized structural recurrent neural networks for crowd navigation using deep reinforcement learning, which has garnered 118 citations and addressed critical limitations in partially observable settings where agent dynamics are unknown. Building on this foundation, he advanced the field further with intention-aware navigation using attention-based interaction graphs (76 citations), enabling robots to model diverse agent interactions and anticipate human intentions in dense crowds. Beyond navigation, Chang has pioneered multimodal robot control, investigating how robots can interpret sound and visual signals together for more natural human-robot communication. His DRAGON system (18 citations) demonstrates a dialogue-based assistive robot for visually impaired individuals, combining visual-language grounding with semantic navigation. With over 200 cumulative citations, Chang's body of work reflects a consistent commitment to making robots more perceptive, adaptive, and genuinely useful in human-centered environments.

Research Focus

Key Achievements

4
H-Index
7
Papers
226
Total Citations
32
Avg Citations/Paper
🏆 Most Cited Paper
Decentralized Structural-RNN for Robot Crowd Navigation with Deep Reinforcement Learning
118 citations · 2021
📈 Most Prolific Year: 2023 (2 Papers)
🤝 Key Collaborators: 14
🏛 Institutions: University of Illinois Urbana-Champaign

Top Papers

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 15 days ago