EEG Based Brain-Controlled Vehicle

M, Raman Kumar · 2026 · Crossref

DOI: 10.46632/jeae/5/2/19

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This paper presents the design and implementation of a low-cost, brain-controlled robotic vehicle that utilizes Electroencephalogram (EEG) signals to enable hands-free operation. The research addresses the limitations of traditional assistive technologies, which often require physical input devices unsuitable for individuals with severe motor impairments, and existing Brain-Computer Interface (BCI) systems, which are frequently complex, expensive, and require extensive user training. The authors propose a simplified approach that leverages eye movement artifacts—typically considered noise in EEG recordings—as distinct control signals, thereby reducing computational complexity and hardware costs while maintaining real-time responsiveness. The system architecture consists of two primary units: a head-worn transmitter and a vehicle-mounted receiver. The transmitter employs Ag/AgCl electrodes placed at the FP1 and FP2 positions on the forehead to capture bio-signals. These weak microvolt-level signals are amplified by a factor of 1000 using an AD620 instrumentation amplifier and filtered to remove high-frequency interference. The conditioned analog signal is digitized by an ESP32 microcontroller’s 12-bit ADC at a 50 Hz sampling rate. Instead of using complex machine learning algorithms, the system applies a single-pole Infinite Impulse Response (IIR) filter to smooth the data and uses threshold-based classification to detect specific voltage ranges corresponding to leftward gaze, rightward gaze, or a neutral state. Control commands are transmitted wirelessly to the receiver unit via the ESP-NOW protocol, a connectionless Wi-Fi layer that ensures sub-millisecond latency without requiring a router. The receiver ESP32 decodes these commands and drives DC motors through an L298N motor driver. Additionally, an HC-SR04 ultrasonic sensor provides obstacle avoidance functionality, overriding motion commands if an object is detected within 20 cm. Experimental results from indoor trials demonstrate the system’s effectiveness and reliability. The study recorded 40 combined trials of deliberate horizontal eye movements, achieving a command recognition accuracy of 91% for left turns and 88% for right turns. The primary source of error was identified as boundary crossings into the neutral zone during slow eye-movement transitions. The end-to-end system latency, measured from eye movement onset to motor actuation, was approximately 120 ms, primarily due to the ADC sampling period and a 1-second refractory timer designed to prevent spurious re-triggering. The ESP-NOW communication link operated reliably within a 10-meter line-of-sight range with zero packet loss, and the ultrasonic sensor successfully halted the vehicle in all obstacle-detection trials. The significance of this work lies in its demonstration of a functional, plug-and-use BCI system that eliminates the need for dedicated Electrooculography (EOG) hardware or complex signal processing pipelines. With an estimated component cost below USD 50, the platform offers an accessible solution for educational purposes and early-stage assistive device research. The authors conclude that this approach simplifies system design and reduces setup complexity compared to traditional motor-imagery or SSVEP-based BCIs. Future enhancements proposed include integrating machine learning classifiers to distinguish a richer set of mental states, adding forward and backward commands triggered by blink patterns, and adapting the chassis for use as a smart wheelchair with caregiver override capabilities.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-nvidia on 2026-08-10; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• drowsiness detection algorithms