Our research in in-sensor computing for artificial robot vision lies at the forefront of integrating advanced sensor architectures with embedded computational capabilities to enhance robotic perception. For example, a curved neuromorphic image sensor array based on MoS₂-organic and IGZO-organic heterostructures (^(ⅰ)Nature Communications, 2020; ^(ⅱ)Science Advances, 2022) demonstrates how bio-inspired design can optimize signal processing, reduce computational overhead, and improve the efficiency of semantic segmentation (^(ⅲ)Science Advances, 2025). In addition, the development of an anti-distortion bio-inspired camera featuring an inhomogeneous photo-pixel array (^(ⅳ)Nature Communications, 2024) introduces innovative strategies for mitigating image distortion in complex environments. AI-native imaging systems inspired by various natural vision systems, such as the cuttlefish (^(ⅴ)Science Robotics, 2023) and hawk (^(M)Science Robotics, 2024) eyes, enable a significant reduction in computational load for object detection, recognition, and tracking. Collectively, these breakthroughs underscore our commitment to pioneering next-generation human-friendly robotic systems.