Hevolve AI: Self-Evolving Multimodal AI Agents

Turn your domain expertise into AI agents that keep learning. Hevolve AI lets experts build multimodal AI systems by talking to them and correcting them in real time, with no code to write.

Key Features

Quick Links

© 2024 Hevolve AI Pvt Ltd. All rights reserved.

← All research
Brain–Computer Interfaces
arXiv (BCI) · June 15, 2026

An Augmented Reality Brain-Robot Interface for Generalist Robot Arm Manipulation

Shangkai Zhang, Rousslan Fernand Julien Dossa, Luca Nunziante +2 more

Controlling a robot arm with EEG has an unavoidable bandwidth problem. Brain signals carry far less information per second than a hand on a joystick, so a system that tries to steer every degree of freedom from the scalp will always feel slow and imprecise.

This design sidesteps that by splitting the job between two channels suited to different parts of it. Gaze picks the object, which is what eyes are naturally good at and what a camera can read easily, and motor imagery drives the manipulation. Neither channel is asked to do work the other does better.

The word doing the most work in the title is generalist. Existing AR brain-robot systems tend to be built around a fixed task structure, which is what makes them demo well and deploy badly, since a real environment does not agree in advance which objects will be present or what should be done with them.

From the arXiv (BCI) abstract

The integration of augmented reality (AR) and EEG-based brain-computer interfaces (BCIs) offers a promising path for enabling intuitive control of robots for assistive purposes. However, existing AR brain-robot interface (BRI) systems are often constrained to task-specific structures, limiting their utility in real-world environments. We present an AR BRI designed for generalist robot arm manipulation that combines gaze-based object selection with motor imagery action…


More Brain–Computer Interfaces papers