Mutual Cognitive Human-Robot Collaborative Manufacturing System

This innovative technology offers a promising mutual-cognitive human-robot collaborative manufacturing system (MCHRCMS), that provides a safer and more intuitive way for humans and robots to work together. Its main features and advantages include:

• Vision-based HRC holistic scene perception. Advanced machine learning techniques accurately predict human intentions, estimate object 6DoF positions, and segment the 3D working environment with an average accuracy rate of over 97%.
• HRC safe interaction strategies. Digital twin and augmented reality (AR) technologies are adopted to ensure human-robot safety and prevent collisions. The system leverages deep reinforcement learning and inverse kinematics to plan robot motions and avoid errors with an accuracy rate of over 99.5%.
• AR deployment of visual reasoning-based cognitive decisions. AR provides intuitive support for HRC instructions with an overall response time of less than 0.6s, making the system friendly to use.

This technology can be well-adopted in human-robot symbiotic manufacturing scenarios under Industry 4.0/5.0.