Abstract: A machine intelligence assistant BCI (brain-computer interface) navigation method for an outdoor mobile robot is put forward in view of the problem of BCI's low signal-to-noise ratio, bad accuracy and long time delay. A vehicle navigation system based on BCI and dual laser radar is designed and implemented. Firstly, an improved angle potential field method based on dual laser radar is used for local path planning, then with navigation intention from BCI system, control commands are generated by fusion decision and used for driving a electric vehicle with modified mechanical system. Experiments show that the system can realize intelligent obstacle avoidance and human-machine collaborative navigation based on environmental obstacle information and brain-machine interface control intention, and it has higher accuracy, tolerance and robustness.