Can Pu

About

Can Pu is a robotics researcher whose work sits at the intersection of legged locomotion, mobile manipulation, and bio-inspired control. Their most impactful contribution addresses a fundamental challenge in legged robotics: fall recovery. In their highly cited 2023 paper (20 citations), Pu developed a framework enabling robots to learn complex motor skills for recovering from falls in unstructured, real-world environments—a critical capability for achieving true all-terrain traversability. This work is complemented by a 2025 paper (6 citations) that draws inspiration from biological central pattern generators (CPGs) in the spinal cord, proposing a novel neural network approach to learn multi-skill locomotion with inherent rhythmicity. Pu has also advanced mobile manipulation, creating a general automation framework for flexible tasks in controlled environments (7 citations), and contributed to environmental robotics with a multi-modal garden dataset and hybrid 3D reconstruction system for trimming robots (4 citations). By bridging bio-inspired control with practical robotic systems, Pu is helping to build more resilient, versatile robots capable of operating beyond laboratory settings.

Research Focus

Key Achievements

Top Papers

1. 1
   Learning Complex Motor Skills for Legged Robot Fall Recovery

   20 citations · 2023
2. 2
   A general mobile manipulator automation framework for flexible tasks in controlled environments

   7 citations · 2023
3. 3
   Bio-inspired neural networks with central pattern generators for learning multi-skill locomotion

   6 citations · 2025
4. 4
   A multi-modal garden dataset and hybrid 3D dense reconstruction framework based on panoramic stereo images for a trimming robot

   4 citations · 2023

Key Collaborators

Contact & Links