Bhaskar Anand

National Institute of Technology, Surat, Electronics Engineering, Graduate Student

Followers

Following

Public Views

Address: India

less

Javier Díaz Noci

Pompeu Fabra University

Steffen Boehm

University of Exeter

David Seamon

Kansas State University

Armando Marques-Guedes

UNL - New University of Lisbon

Fatiha Djemili Tolba

Badji Mokhtar - Annaba University

Roshan Chitrakar

Nepal College of Information Technology

J.-C. Spender

Kozminski University

Lev Manovich

Graduate Center of the City University of New York

Isad Saric

University of Sarajevo

Journal of Computer Science IJCSIS

Interests

Uploads

Papers by Bhaskar Anand

Object Detection and Segmentation using LiDAR-Camera Fusion for Autonomous Vehicle

2021 Fifth IEEE International Conference on Robotic Computing (IRC)

Quantitative Comparison of LiDAR Point Cloud Segmentation for Autonomous Vehicles

2021 IEEE 94th Vehicular Technology Conference (VTC2021-Fall)

An experimental analysis of various multi-channel LiDAR systems

2020 IEEE International Conference on Computing, Power and Communication Technologies (GUCON), 2020

Depth perception is essential for applications like autonomous vehicles, robotics, and automation... more Depth perception is essential for applications like autonomous vehicles, robotics, and automation. Stereo vision camera provides one way for the above purpose but has a limited range. Radar has also been used for depth sensing but does not provide precise resolution. LiDAR (Light Detection and Ranging) provides a three-dimensional view of a scene with much higher resolution as compared to Radar and has a much larger range than Stereo vision camera. In this paper, we present a complete overview of LiDAR, which includes basic principles, data acquisition methods, a comparative study among different sensors. An analysis of variation in data rate and the number of points for different return modes with the varying number of channels is also presented.

Comparative Run Time Analysis of LiDAR Point Cloud Processing with GPU and CPU

The Light detection and ranging (LiDAR) sensor is well known for its application in the field of ... more The Light detection and ranging (LiDAR) sensor is well known for its application in the field of surveying and autonomous vehicles. It is used for capturing the three-dimensional view of a scene. The output of the LiDAR sensor, known as point cloud, contains a large number of spatial points leading to bulky data. Any algorithm developed for this data consumes too much time, making it impractical to be used for real-time applications. Recently, Graphics Processing Units (GPUs) have been extensively used for performing the parallel operation. In this paper, we have quantitatively verified the usefulness of GPU for faster LiDAR point cloud data processing as compared to CPU. Algebraic operations were performed on LiDAR point cloud data on GPU, and the run time was found to decrease up to one-sixth as compared to Central Processing Unit (CPU).

Fast Object Segmentation Pipeline for Point Clouds Using Robot Operating System

2019 IEEE 5th World Forum on Internet of Things (WF-IoT)

This paper presents a method to pipeline the segmentation process for point clouds using the Robo... more This paper presents a method to pipeline the segmentation process for point clouds using the Robot Operating System (ROS) and the Point Cloud Library (PCL). The pipeline's objective is to optimize the run time of a conventional segmentation algorithm by working within the Robot Operating System framework. It can be implemented using any system and in conjunction with a GPU. It shows the greatest reduction in run time for the least downsampled clouds. Therefore, it can be used for real-time safety-critical applications especially in scenarios where the point cloud is sparse or of highly uneven spatial density and should thus not be downsampled. It was developed for Obstacle Avoidance for Autonomous Vehicles and Drones where segmentation is only the first step of a larger pipeline involving obstacle detection and tracking. It was observed to reduce run time up to 31.3% on the KITTI data set and up to 44.4% on data collected from a 16 channel Ouster lidar at the

Download

LiDAR-INS/GNSS Based Real-Time Ground Removal, Segmentation and Georeferencing Framework for Smart Transportation

IEEE Transactions on Instrumentation and Measurement

Light detection and ranging (LiDAR) sensor is attracting significant attention in the field of sm... more Light detection and ranging (LiDAR) sensor is attracting significant attention in the field of smart transportation because of its ability to give depth information. It is already extensively used for obstacle detection. The inertial navigation system (INS) with a global navigation satellite system (GNSS) gives the global position, orientation, and velocity of a given object in real time. In autonomous driving, LiDAR point clouds are segmented to find out the positions of various objects in the surrounding of the vehicle of interest. However, the obstacle position is determined in local east–north–up (ENU) coordinates. In this article, an end-to-end framework has been presented, which takes input from LiDAR and INS/GNSS systems and gives the obstacle position in universal coordinates (latitude, longitude) in real time. The proposed framework includes ground point removal from raw LiDAR data, obstacle segmentation, and LiDAR data fusion with INS/GNSS data for georeferencing. For LiDAR data processing, a novel method for removing ground points has been presented, in which the entire point cloud is divided into square grids in the horizontal plane. Based on the statistics of the vertical distribution of the points in each grid, ground points are identified. Two ground point datasets were also created for testing the proposed algorithm. These datasets contain point clouds of ground with varying inclination from various multichannel LiDARs. The proposed ground removal method was tested on the Paris-Lille-3D dataset and the dataset created. An F1-score of greater than 0.99 and 0.98 was achieved on our dataset and the Paris Lille-3D dataset, respectively. The framework is tested on different hardware configurations and found to be suitable for real-time applications.

Geo-referencing system for locating objects globally in LiDAR point cloud

2020 IEEE 6th World Forum on Internet of Things (WF-IoT)

The LiDAR (Light detection and ranging) sensor is widely used to collect a three-dimensional view... more The LiDAR (Light detection and ranging) sensor is widely used to collect a three-dimensional view of the surrounding. This LiDAR’s data is represented in the local East North Up (ENU) coordinate system and does not contain any global information. Converting this data into the global coordinate system will be helpful in intelligent transportation system (ITS) for object location tracking and sharing. This type of conversion is commonly known as geo-referencing. Generally, geo-referencing is used to associate a physical map or raster image with global locations in the field of Geographical Information System (GIS). In this paper, we have suggested a system for geo-referencing LiDAR point cloud, global positioning of objects present in the point cloud and sharing it with other communicating units present in the vehicular network. We have tested our system on data captured at the Indian Institute of Technology Hyderabad (IITH) campus and compared it with the manually collected GPS location of the objects. Our system is found to determine the global location of the objects with a sufficiently small error percentage of the order of 10−4.