A Multi-perception fusion using shared-control method for brain-mobile robot

Area of Science:

• Robotics
• Human-Computer Interaction
• Neuroscience

Background:

• Brain-computer interfaces (BCIs) are crucial for enhancing human-robot collaboration adaptability in complex settings.
• Integrating human perception with robot capabilities is key for advanced autonomous systems.

Purpose of the Study:

• To propose and evaluate a multi-perception fusion using shared control (MPF-SC) method for brain-controlled mobile robots.
• To improve navigation and obstacle avoidance in complex terrains by fusing human and robot perceptions.

Main Methods:

• Developed an MPF-SC method integrating electroencephalography (EEG) and electromyography (EMG) signals with user intent.
• Utilized computer vision and a grid costmap to map human perception and dynamically adjust robot navigation.
• Applied the method to brain-controlled mobile robots for navigation and obstacle avoidance tasks.

Main Results:

• MPF-SC generated smoother trajectories and significantly reduced collision rates compared to traditional methods.
• User comfort was significantly enhanced during navigation tasks.
• The method effectively integrated human anticipation of risks with robot environmental perception.

Conclusions:

• The MPF-SC method offers a novel and intuitive approach for human-machine shared control in complex environments.
• Bilateral intelligence, leveraging both human and robot capabilities, enhances adaptability in sophisticated operational settings.
• This approach opens new avenues for adaptive and intuitive human-robot collaboration.