Haoze Zheng

University at Buffalo, State University of New York

Papers

Total Citations

H-Index

About

Haoze Zheng is a robotics researcher whose work lies at the intersection of perception, planning, and control, with a particular focus on kinematics-aware trajectory generation for autonomous systems. His most-cited paper, "iKap: Kinematics-Aware Planning with Imperative Learning" (2025), introduces a novel framework that bridges vision-based environmental understanding with motion planning. By integrating imperative learning—a paradigm that allows planning modules to adapt through differentiable constraints—Zheng’s approach enables robots to generate collision-free, executable pose sequences that account for kinematic feasibility in real time. This work addresses a critical bottleneck in modular vision-to-planning pipelines, where traditional systems often fail to adapt to dynamic surroundings or enforce motion constraints. Though early in its citation impact (2 citations), iKap represents a significant step toward more resilient and interpretable robotic systems. Zheng’s contributions are particularly relevant for applications in autonomous navigation, manipulation, and human-robot interaction, where reliability and adaptability are paramount. His research exemplifies a growing trend toward end-to-end differentiable planning, promising to make robots not just perceptive, but truly responsive to their physical constraints.