Parag Sewalkar

Mr. Parag Sewalkar

orcid

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
17 January 2019 in Sensors
Verified
Journal Article
Open access
PDF file
PSParag Sewalkar
Parag Sewalkar
JSJochen Seitz
Jochen Seitz
In the last few years, increasing attention has been provided to research Vehicle-to-Pedestrian (V2P) communication systems. These V2P systems serve different purposes (safety or convenience) and cater to different Vulnerable Road User (VRU) groups. Also, these V2P systems employ different communication technologies, and use different mechanisms to interact with the users. An effective V2P system also needs to consider varying characteristics of different VRUs. These various elements may be considered as design parameters of the V2P system. In this paper, we discuss such elements and propose a design framework for the V2P system based on them. We also provide an extensive survey of existing V2P efforts for safety and convenience applications and their design considerations. We perform a case study that compares the different approaches of V2P safety system for different VRU groups under different pre-crash scenarios. Finally, we discuss a few technological challenges in integration of VRUs into V2X systems.

22 December 2020 in IEEE Transactions on Intelligent Vehicles
Verified
Journal Article
Closed access
PSParag Sewalkar
Parag Sewalkar
JSJochen Seitz
Jochen Seitz
Vehicle-to-Pedestrian communication can extend crash prevention capabilities of the current driver assistance systems in vehicles. This requires vehicles and pedestrians to exchange safety messages with each other. However, as the number of pedestrians increases, the numerous safety messages transmitted by pedestrians can quickly congest the network. This can severely affect Vehicle-to-Pedestrian and Vehicle-to-Vehicle communication. Hence, a mechanism for the mitigation of network congestion caused by pedestrian safety messages is required. This paper proposes a Multi-channel Clustering-based Congestion Control (MC-COCO4V2P) algorithm, a proactive and infrastructure-independent clustering-based approach to mitigate the network congestion caused by pedestrians. Our approach clusters pedestrians based on their location and direction and uses separate channels for exchanging cluster and safety messages, thereby reducing the control information overhead. It also employs a transmit power control mechanism to make the clustering mechanism energy efficient. Our results show that the clustering of pedestrians can significantly improve network performance and reduce the power consumption of pedestrians' devices.