Max Robotics — Global Service Robot Directory + US Certification

About

Zongzheng Zhang is a rising researcher at the forefront of embodied AI and autonomous driving perception, whose work bridges the gap between high-level reasoning and low-level robotic control. His research centers on neuro-symbolic reasoning for scene understanding, affordance-based manipulation, and part-level dynamics modeling. In his influential paper "Chameleon: Fast-Slow Neuro-Symbolic Lane Topology Extraction" (2025, 5 citations), Zhang pioneered a dual-speed reasoning framework that enables autonomous vehicles to perform complex relational reasoning—such as determining lane-change feasibility—a critical step toward mapless driving. His work "PreAfford: Universal Affordance-Based Pre-Grasping for Diverse Objects and Environments" (2024, 4 citations) introduced a novel approach to robotic grasping that allows two-finger grippers to manipulate objects lacking distinct features, dramatically expanding the range of manipulable items without requiring external aids. Most recently, in "PartRM: Modeling Part-Level Dynamics with Large Cross-State Reconstruction Model" (2025, 2 citations), Zhang addresses the challenge of predicting future object states by modeling part-level dynamics, advancing the development of world models for robotics. Despite the early stage of his career, Zhang’s work has already garnered attention for its innovative integration of symbolic reasoning with neural networks, positioning him as a promising voice in the next generation of AI researchers tackling real-world perception and manipulation challenges.