Abstract: Autonomous vehicle systems provide improved safety in comparison to human driven vehicles by reducing crashes. However, in future scenarios, the autonomous vehicles have to share the road with human-driven vehicles. But human drivers make decision errors during maneuvers which lead to crashes. Autonomous vehicles therefore have to predict human-initiated safe and unsafe maneuvers to react appropriately for the safety and comfort of its passengers. The present disclosure addresses unresolved problems of the conventional autonomous vehicle systems by providing a stacked autoencoder based singular deep neural network that uses sensor data to predict if any vehicle within its field of view would perform safe or unsafe lane maneuvers into its headway or into any other vehicle's headway. The method of the present disclosure also helps surrounding vehicles to know intention of a target vehicle with the help of vehicle-to-vehicle communication systems in a road driving environment.

Abstract: State of the art approaches for training chained neural network models for delay prediction train the data models using only real data and not predicted data. Such models when used in a chained way leads to worse results as they are not exposed to predicted data during training. This leads to the model prediction errors showing sharp increase as the models tries to predict for subsequent stations past the immediate station. Embodiments disclosed herein provide a method and system for training of chained neural networks for delay prediction in transit networks. In this approach, a chained neural network model used by the system is trained such that data containing a mix of real data and predicted data is used for training each data model in a sequence of data models in the chained neural network model.

Abstract: The present disclosure provides a system and method for delay prediction for scheduled public transport. A multi-architectural deep learning approach has been used to predict the delays of a queried vehicle in the scheduled public transport. For this, historical operational data is transformed into temporal, and spatiotemporal data. While, the spatial data is obtained from geographical information. The system uses different combinations of neural networks architectures. A regressor model uses three separate kinds of architecture. One component is the Fully Connected Neural Network (FCNN), which is good at learning from static features, the second is the Long Short Term Memory (LSTM) network which is good at learning from temporal features, and the third is the 3D Convolutional Neural Network (3DCNN) which is good at learning from spatiotemporal features. Learned encoding from each are fed to another FCNN to produce the predicted delay value.

Abstract: The present disclosure predicts a delay associated with a vehicle. Conventional methods are mainly mathematical based and machine learning based networks are not predicting delay accurately. Initially, the present disclosure Initially, the system receives a user query comprising an expected delay of a target vehicle in at least one target station. Further, a real time data associated with the user query in a predefined horizon is obtained. Further, a spatial feature vector, a temporal feature vector and spatiotemporal features are extracted based on the real time data using a feature extraction technique. Finally, the expected is predicted based on the plurality of features using a trained adversarial regression model, wherein the trained adversarial regression model comprises a critic network and a regressor network. The regressor network is trained with a plurality of architectures and a best architecture with minimum Mean Absolute Error (MAE) is selected for delay prediction.

Abstract: Vehicle-to-Grid (V2G) technologies are being adopted to reduce peak demand and to take over as energy sources during grid instability. It is necessary to estimate attributes of electric vehicle trips and residual battery charge in order to correctly predict spatio-temporal availability of energy from EVs to form a micro grid. However, it may not be feasible to get the required attributes for all vehicles in a city-scale traffic scenario. Embodiments of the present disclosure and system address the problem of accurately estimating the local energy reserve that is available from parked EVs during a given time of the day. In addition, the system also determines which neighborhoods have the potential to form micro grids using the parked EVs during a given time period. This will help grid operator(s) to plan and design smart grids which can create EV-powered micro grids in neighborhoods during periods of peak demand or during disruptions.

Abstract: Method and system for predicting temporal-spatial distribution of load demand on an electric grid due to a plurality of Electric Vehicles (EVs) is described. The method includes creating an EV load demand (EVLD) model for a Region of Interest (ROI) serviced by the electric grid, wherein the EVLD model integrates an EV model and a transport simulator simulating EV traffic conditions for the ROI. Further, the method includes computing the load demand in time and space in terms of State of Charge (SOC) of a battery for each EV among the plurality of EVs in the ROI, based on the EVLD model. Furthermore, the method includes aggregating the computed the load demand, in terms of the SOC, of each EV in time domain and space domain to create a temporal-spatial impact of the load demand by the plurality of EVs on the electric grid for the ROI.

Abstract: Method and system for predicting temporal-spatial distribution of load demand on an electric grid due to a plurality of Electric Vehicles (EVs) is described. The method includes creating an EV load demand (EVLD) model for a Region of Interest (ROI) serviced by the electric grid, wherein the EVLD model integrates an EV model and a transport simulator simulating EV traffic conditions for the ROI. Further, the method includes computing the load demand in time and space in terms of State of Charge (SOC) of a battery for each EV among the plurality of EVs in the ROI, based on the EVLD model. Furthermore, the method includes aggregating the computed the load demand, in terms of the SOC, of each EV in time domain and space domain to create a temporal-spatial impact of the load demand by the plurality of EVs on the electric grid for the ROI.

Abstract: Vehicle-to-Grid (V2G) technologies are being adopted to reduce peak demand and to take over as energy sources during grid instability. It is necessary to estimate attributes of electric vehicle trips and residual battery charge in order to correctly predict spatio-temporal availability of energy from EVs to form a micro grid. However, it may not be feasible to get the required attributes for all vehicles in a city-scale traffic scenario. Embodiments of the present disclosure and system address the problem of accurately estimating the local energy reserve that is available from parked EVs during a given time of the day. In addition, the system also determines which neighborhoods have the potential to form micro grids using the parked EVs during a given time period. This will help grid operator(s) to plan and design smart grids which can create EV-powered micro grids in neighborhoods during periods of peak demand or during disruptions.