Autonomous Motion Vision with Tri-bulk-heterojunctioned Organic Adaptation Transistor

Area of Science:

• Materials Science
• Optoelectronics
• Robotics and Autonomous Systems

Background:

• Motion vision is crucial for autonomous driving and embodied intelligence.
• Current sensors separate motion analysis and moving-object extraction, increasing computational and storage demands.
• This inefficiency challenges the development of highly efficient real-world applications.

Purpose of the Study:

• To design a single device integrating motion analysis and moving-object extraction.
• To develop an organic phototransistor (OTAT) for efficient in-sensor motion vision.
• To overcome the limitations of separate sensor functionalities.

Main Methods:

• Designed an organic tri-bulk-heterojunctioned adaptation phototransistor (OTAT).
• Incorporated a gate-adjacent semiconductor-insulator heterojunction as a photosensitive percolation channel.
• Conducted imaging experiments to evaluate in-sensor capabilities.

Main Results:

• The OTAT achieved in-sensor moving-object extraction with a 45.3% highlight ratio.
• Demonstrated motion analysis accuracy exceeding 99.4% for direction and speed detection.
• OTAT-based agents showed the lowest risk in route-planning simulations.

Conclusions:

• The OTAT effectively integrates motion extraction and analysis into a single device.
• This represents a significant advancement for next-generation in-sensor motion vision systems.
• The technology shows promise for improving efficiency in autonomous applications.