Disciplines
Civil Engineering | Electrical and Computer Engineering | Systems and Communications | Transportation Engineering
Abstract (300 words maximum)
In the year 2019, more than 6,500 fatal crashes on the United States roadways involved pedestrians. Furthermore, despite the significant reduction in vehicle miles travel during 2020, pedestrian deaths have increased by 4.8%. Vehicle-to-Pedestrian (V2P) communication systems is one of the solutions to reduce pedestrian fatalities and improve safety. This system improves pedestrian safety by alerting either a vehicle or pedestrian of a potential collision in real-time. Performance of this system can be significantly enhanced by predicting movement of pedestrians, specifically estimating the intent to cross a roadway; and broadcasting this information in real-time using various communication protocols. We aim on understanding and analyzing the natural fluctuations of the cognitive state. Understanding these fluctuations can lead to the development of a predictive method to detect the change in the cognitive state and a V2P system can be designed to broadcast the change to nearby vehicles. In the first phase of this study, the focus is to gain new insights about pedestrian and automated driver communication techniques for a safe traffic environment. We have developed a predictive methodology to detect the intention to start movement with the use of electroencephalography (EEG) signals from the brain. For this study, subjects will be monitored using the OpenBCI Ultracortex software as they cross a road. The EEG signals acquired will be processed and analyzed. With this information an algorithm will be designed to anticipate pedestrian movement which can be disseminated to nearby vehicles using various communication techniques.
Academic department under which the project should be listed
SPCEET - Electrical and Computer Engineering
Primary Investigator (PI) Name
Sylvia Bhattacharya
Additional Faculty
Parth Bhavsar, Civil Engineering, pbhavsar@kennesaw.edu Scott Larisch, Electrical Engineering, slarisc1@kennesaw.edu
Prediction of Gait Intention from Pre-movement EEG Signals for Vehicle-to-Pedestrian (V2P) Communication Modeling
In the year 2019, more than 6,500 fatal crashes on the United States roadways involved pedestrians. Furthermore, despite the significant reduction in vehicle miles travel during 2020, pedestrian deaths have increased by 4.8%. Vehicle-to-Pedestrian (V2P) communication systems is one of the solutions to reduce pedestrian fatalities and improve safety. This system improves pedestrian safety by alerting either a vehicle or pedestrian of a potential collision in real-time. Performance of this system can be significantly enhanced by predicting movement of pedestrians, specifically estimating the intent to cross a roadway; and broadcasting this information in real-time using various communication protocols. We aim on understanding and analyzing the natural fluctuations of the cognitive state. Understanding these fluctuations can lead to the development of a predictive method to detect the change in the cognitive state and a V2P system can be designed to broadcast the change to nearby vehicles. In the first phase of this study, the focus is to gain new insights about pedestrian and automated driver communication techniques for a safe traffic environment. We have developed a predictive methodology to detect the intention to start movement with the use of electroencephalography (EEG) signals from the brain. For this study, subjects will be monitored using the OpenBCI Ultracortex software as they cross a road. The EEG signals acquired will be processed and analyzed. With this information an algorithm will be designed to anticipate pedestrian movement which can be disseminated to nearby vehicles using various communication techniques.