Neurobotics: Merging Neuroscience and Robotics with Yoky Matsuoka

Introducing Neurobotics

• 💡 Neurobotics is a new field merging neuroscience and robotics, created about 10 years ago to help people.
• 🎯 The core idea is to enable individuals with neurological disorders or physical disabilities to control robotic devices using their brain signals.
• 🧠 An example is allowing paralyzed individuals to control robotic wearable devices with their thoughts, mimicking natural limb movement.

The Genesis of an Idea

• 🎾 The speaker's journey began as a passionate tennis player who experienced frequent injuries, leading her to seek a solution.
• 🤖 Initially, she aimed to build a human-like tennis buddy robot with human-level intelligence and capabilities for playing tennis.
• 🔬 Early work involved building robotic legs at UC Berkeley and replicating a two-year-old's intelligence in a robot's arms and hands at MIT, achieving human-like reflexes.

Shifting Focus to Neuroscience

• ⚠️ Realizing the limitations of robotics alone for achieving true human dexterity, the speaker decided to study neuroscience at MIT.
• 🧠 Understanding the human brain's complexity became crucial for improving robotic systems and their control.
• 🤝 The new goal became combining neuroscience and robotics to help people with neurological disorders or physical disabilities regain movement.

Current Progress and Challenges

• 🔬 In human trials, primitive brain-computer interfaces allow locked-in patients to control a computer cursor for basic communication (e.g., Cyber Kinetics).
• 🐒 Animal studies demonstrate that monkeys can control robotic arms with brain signals to reach for food, though fine hand dexterity is still limited.
• 🦾 Prosthetic advancements include muscle nerve reinervation for amputees to control robotic arms, but achieving sophisticated hand movements remains a significant challenge.

Advancing Robotic Dexterity

• 🛠️ Traditional robotic hands often have mechanical limitations, such as rigid palms, which hinder human-like dexterity despite advanced designs.
• 🦴 The speaker's team adopted an anatomical approach, dissecting human hands to build robotic systems that precisely mimic bone structure and muscle-tendon connections.
• 🚀 This anatomical robot can be controlled by human finger movements, muscle signals, or neural signals, serving as a testbed for understanding brain signals and achieving advanced dexterity.

Broader Impact of Neurobotics

• 💡 Beyond prosthetics, neurobotics encompasses exoskeletons for augmentation (e.g., military applications) and rehabilitation (e.g., for stroke patients).
• 🌐 The field also explores immersive environments and direct brain-neuron-silicon interfaces for various applications.
• 🌱 Neurobotics is inspiring a new generation of engineers and scientists by demonstrating the human-centric and impactful applications of science and technology.