Users can now command a robot dog using nothing but their mental intent, as a new system allows the machine to autonomously plan its path, avoid obstacles and navigate to a designated location with a thought.
The development was achieved at Xi’an Jiaotong University in northwest China, where Professor Xu Guanghua and his team successfully integrated electroencephalogram (EEG)-based control with autonomous navigation.
At the core of the breakthrough is non-invasive brain-computer interface (BCI) technology, which captures electrical signals from neuronal activity to enable precise control of mechanical devices. Xu described the system as a form of “remote control in your mind,” where user intentions are decoded from brain signals and translated into commands for the robot.
When a user thinks of a command such as moving forward, the brain generates corresponding EEG signals. The system collects and interprets these signals, identifies the intended action, and sends instructions to the robot dog, which then executes the movement autonomously.
Currently, the system supports 11 basic mental commands, including forward movement, backward motion and turning. It reportedly achieves a recognition accuracy of over 95 percent, with a response delay of approximately one second between thought and action.
The project reflects growing global interest in brain-computer interface technologies. While invasive methods offer higher precision, they require surgical implantation and carry risks such as infection and immune rejection. In contrast, non-invasive approaches are safer, more cost-effective and easier to scale, making them suitable for applications in rehabilitation and consumer use.
However, non-invasive signals are inherently less precise, posing challenges for continuous real-time control. To overcome this limitation, the research team adopted a human-machine collaboration model, assigning humans high-level decision-making tasks while delegating complex operations such as navigation, obstacle avoidance and motion execution to the machine’s autonomous systems.
Xu said this division of roles improves efficiency and stability, combining human intent with machine precision. He emphasised that future progress in BCI technology will depend on integrating advancements in artificial intelligence, autonomous navigation and intelligent perception.
The team envisions broader applications of such systems in assisting individuals with disabilities, elderly care, medical support, rehabilitation training and intelligent companionship, potentially transforming robots into practical everyday assistants.