Arnob Ghosh

Arnob Ghosh

Auto-generated profile, based on publicly available information

Arnob Ghosh

New Jersey Institute of Technology

Research Focus & Metrics

Academic Career Length

17Years (2010-Present)

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

Representative Publications

View All Publications
9 August 2019 in IEEE Transactions on Intelligent Transportation Systems
Verified
Journal Article
Closed access
AAAbubakr O. Al-Abbasi
Abubakr O. Al-Abbasi
AGArnob Ghosh
Arnob Ghosh
Vaneet AggarwalVaneet Aggarwal
Vaneet Aggarwal
The success of modern ride-sharing platforms crucially depends on the profit of the ride-sharing fleet operating companies, and how efficiently the resources are managed. Further, ride-sharing allows sharing costs and, hence, reduces the congestion and emission by making better use of vehicle capacities. In this paper, we develop a distributed model-free, DeepPool, that uses deep Q-network (DQN) techniques to learn optimal dispatch policies by interacting with the environment. Further, DeepPool efficiently incorporates travel demand statistics and deep learning models to manage dispatching vehicles for improved ride sharing services. Using real-world dataset of taxi trip records in New York, DeepPool performs better than other strategies, proposed in the literature, that do not consider ride sharing or do not dispatch the vehicles to regions where the future demand is anticipated. Finally, DeepPool can adapt rapidly to dynamic environments since it is implemented in a distributed manner in which each vehicle solves its own DQN individually without coordination.

28 April 2017 in IEEE Transactions on Smart Grid
Verified
Journal Article
Closed access
AGArnob Ghosh
Arnob Ghosh
Vaneet AggarwalVaneet Aggarwal
Vaneet Aggarwal
We propose an online pricing mechanism for electric vehicle (EV) charging. A charging station decides prices for each arriving EV depending on the energy and the time within which the EV will be served (i.e., deadline). The user selects either one of the contracts by paying the prescribed price or rejects all depending on their surpluses. The charging station can serve users using renewable energy and conventional energy. Users may select longer deadlines as they may have to pay less because of the less amount of conventional energy, however, they have to wait a longer period. We consider a myopic charging station and show that there exists a pricing mechanism which jointly maximizes the social welfare and the profit of the charging station when the charging station knows the utilities of the users. However, when the charging station does not know the utilities of the users, the social welfare pricing strategy may not maximize the expected profit of the charging station and even the profit may be 0. We propose a fixed profit pricing strategy which provides a guaranteed fixed profit to the charging station and can maximize the profit in practice. We empirically show that our proposed mechanism reduces the peak-demand and utilizes the limited charging spots in a charging station efficiently.

15 August 2018 in IEEE Transactions on Vehicular Technology
Verified
Journal Article
Closed access
AGArnob Ghosh
Arnob Ghosh
Vaneet AggarwalVaneet Aggarwal
Vaneet Aggarwal
The paper considers a bidirectional power flow model of the electric vehicles (EVs) in a charging station. The EVs can inject energy by discharging via a vehicle-to-grid (V2G) service, which can enhance the profits of the charging station. However, frequent charging and discharging degrade battery life. We propose a menu-based pricing scheme, where the charging station selects a price for each arriving user (EV owner) for the amount of battery utilization, the total energy, and the time (deadline) that the EV will stay. The user can accept one of the menus or rejects all. The user reaches its decision based on the utilities. The charging utilizes its own limited renewable energy, and the conventional energy bought from the grid to serve the users. We show that there exists a prior-free pricing mechanism, which maximizes the ex-post social welfare for a myopic scenario. It selects a lower price for the menu containing a higher V2G service. However, the prior free pricing mechanism does not necessarily maximizes the expected profit. We show that the pricing strategy, which sets the price for a menu at a fixed value greater than the marginal cost of serving the menu, maximizes the expected profit for a wide class of utility functions. In the menu-based pricing, when the harvested renewable energy is small the users have higher incentives for the V2G service. We, numerically, show that the charging station's profit and the user's surplus both increase as V2G service is efficiently utilized by the pricing mechanism.

25 July 2024 in IEEE Transactions on Industrial Informatics
Verified
Journal Article
Closed access
UQUbaid Qureshi
Ubaid Qureshi
AGArnob Ghosh
Arnob Ghosh
BPBijaya Ketan Panigrahi
Bijaya Ketan Panigrahi
This article presents real-time Pareto-optimal scheduling for bidirectional electric vehicle (EV) charging in a commercial charging station with on-site renewable energy and battery energy storage to optimize several objectives. To incorporate the inherent uncertainty in the model, mixture density neural networks are presented to estimate the parameters of the probability distribution of demands and deadlines using a negative-log-likelihood loss function. From the joint distribution of demands and deadlines, future EV charging requests are estimated. Furthermore, we formulate the control problem as a multiobjective stochastic convex optimization problem from the perspective of the charging station operator, which simultaneously aims to minimize the total cost of charging, frequent change in charging rates, maximum demand of the charging station and battery degradation costs subject to various system constraints. We empirically evaluate the proposed scheduling policy for optimality gap, competitive ratio, and robustness, and show that the proposed scheduling policy reduces cost by about $30 \%$ over the benchmark scheduling policies.

27 April 2022 in IEEE Transactions on Intelligent Transportation Systems
Verified
Journal Article
Closed access
AGArnob Ghosh
Arnob Ghosh
TPThomas Parisini
Thomas Parisini
We consider an intersection zone where autonomous vehicles (AVs) and human-driven vehicles (HDVs) can be simulteneously present. As a new vehicle arrives, the traffic controller needs to decide and suggest an optimal sequence of the vehicles which will exit the intersection zone. The traffic controller can inform the time at which an AV can cross the intersection; however, the traffic controller can not communicate with the HDVs, rather the HDVs can only be controlled using the traffic lights. We formulate the problem as an integer constrained nonlinear optimization problem. Since the number of possible combinations increases exponentially with the number of vehicles in the traffic system, we relax the original problem and proposes an algorithm which gives the optimal solution of the relaxed problem and yet only scales linearly with the number of vehicles in the system. The numerical validation shows that our algorithm outperforms the First-In-First-Out (FIFO) algorithm.

23 January 2018 in IEEE Transactions on Network Science and Engineering
Verified
Journal Article
Closed access
AGArnob Ghosh
Arnob Ghosh
SSSaswati Sarkar
Saswati Sarkar
The economic model of the Internet of Things (IoT) consists of end users, advertisers and three different kinds of providers-IoT service provider (IoTSP), Wireless service provider (WSP) and cloud service provider (CSP). We investigate three different kinds of interactions among the providers. First, we consider that the IoTSP prices a bundled service to the end-users, and the WSP and CSP pay the IoTSP (push model). Next, we consider the model where the end-users independently pay each provider (pull model). Finally, we consider a hybrid model of the above two where the IoTSP and WSP quote their prices to the end-users, but the CSP quotes its price to the IoTSP.