Dr. Lalitha Dabbiru Profile

Dr. Lalitha Dabbiru

at Mississippi State Univ

SPIE Involvement:

Author

Publications (8)

Proceedings Article | 14 June 2023 Presentation + Paper

Fidelity requirements for simulating sensor performance in autonomous ground vehicles

Chris Goodin, Daniel Carruth, Lalitha Dabbiru, Michael Hedrick, Zachary Aspin, Justin Carrillo, John Kaniarz

Proceedings Volume 12529, 1252908 (2023) https://doi.org/10.1117/12.2661663

KEYWORDS: Cameras, Data modeling, Image classification, Education and training, Device simulation, Sensors, Autonomous vehicles, Computer simulations, Light sources and illumination

Read Abstract +

Proceedings Article | 13 June 2023 Presentation + Paper

Object detection in synthetic aerial imagery using deep learning

Lalitha Dabbiru, Chris Goodin, Daniel Carruth, Jonathan Boone

Proceedings Volume 12540, 1254002 (2023) https://doi.org/10.1117/12.2662426

KEYWORDS: Object detection, Data modeling, Education and training, Airborne remote sensing, Machine learning, Computer simulations, Buildings, Deep learning, Image quality

Read Abstract +

Proceedings Article | 13 June 2023 Presentation + Paper

Analysis of LiDAR configurations on off-road semantic segmentation performance

Jinhee Yu, Jingdao Chen, Lalitha Dabbiru, Christopher Goodin

Proceedings Volume 12540, 1254003 (2023) https://doi.org/10.1117/12.2663098

KEYWORDS: LIDAR, Sensors, Semantics, Computer simulations, Point clouds, Machine learning

Read Abstract +

LiDAR-based 3D semantic segmentation is one of the most widely used perception methods to support scene understanding of self-driving vehicles. Most publicly available LiDAR datasets for driving scene segmentation, such as SemanticKITTI, nuScenes, and SemanticPOSS, provide only a single type of LiDAR configuration. Therefore, testing a trained model with a different channel configuration than the training dataset is sometimes inevitable in real-world applications. Despite the significance of this LiDAR channel mismatch problem in the machine learning pipeline, little research has focused on investigating the impact of the LiDAR configuration shift on a model’s test performance. This paper aims to provide referenceable baseline experiments for the LiDAR configuration shifts. We explore the effect of using different LiDAR channels when training and testing a 3D LiDAR point cloud semantic segmentation model, utilizing Cylinder3D for the experiments. A Cylinder3D model is trained and tested on simulated 3D LiDAR point cloud datasets created using the Mississippi State University Autonomous Vehicle Simulator (MAVS) and 32, 64 channel 3D LiDAR point clouds of the RELLIS-3D dataset collected in a real-world off-road environment. Our experimental results demonstrate that sensor and spatial domain shifts significantly impact the performance of LiDAR-based semantic segmentation models. In the absence of spatial domain changes between training and testing, models trained and tested on the same sensor type generally exhibited better performance. Moreover, higher-resolution sensors showed improved performance compared to those with lower-resolution ones. However, results varied when spatial domain changes were present. In some cases, the advantage of a sensor’s higher resolution led to better performance both with and without sensor domain shifts. In other instances, the higher resolution resulted in overfitting within a specific domain, causing a lack of generalization capability and decreased performance when tested on data with different sensor configurations.

Proceedings Article | 6 June 2022 Presentation + Paper

Simulation-based testing of autonomous ground vehicles

Christopher Goodin, Daniel Carruth, Lalitha Dabbiru, Christopher Hudson, Lucas Cagle, Nicklaus Scherrer, Marc Moore, Paramsothy Jayakumar

Proceedings Volume 12115, 121150J (2022) https://doi.org/10.1117/12.2620502

KEYWORDS: LIDAR, Unmanned vehicles, Sensors, 3D modeling, Reliability, Computer simulations, Systems modeling, Computer architecture, Detection and tracking algorithms

Read Abstract +

Proceedings Article | 12 April 2021 Presentation + Paper

Traversability mapping in off-road environment using semantic segmentation

Lalitha Dabbiru, Suvash Sharma, Chris Goodin, Sam Ozier, Christopher Hudson, Daniel Carruth, Matthew Doude, George Mason, John Ball

Proceedings Volume 11748, 117480C (2021) https://doi.org/10.1117/12.2587661

KEYWORDS: Algorithm development, Convolutional neural networks, Cameras, Stereoscopic cameras, Machine learning, LIDAR, Imaging systems, Evolutionary algorithms, Data centers

Read Abstract +

Proceedings Article | 12 April 2021 Presentation + Paper

Fast terrain traversability estimation with terrestrial lidar in off-road autonomous navigation

Christopher Goodin, Lalitha Dabbiru, Christopher Hudson, George Mason, Daniel Carruth, Matt Doude

Proceedings Volume 11758, 117580O (2021) https://doi.org/10.1117/12.2585797

KEYWORDS: LIDAR, Computer simulations, Algorithm development, Unmanned vehicles, Roads, Micro unmanned aerial vehicles, Environmental sensing, Binary data

Read Abstract +

Proceedings Article | 22 April 2020 Presentation + Paper

Performance analysis of semantic segmentation algorithms trained with JPEG compressed datasets

Suvash Sharma, Christopher Hudson, Daniel Carruth, Matt Doude, John Ball, Bo Tang, Chris Goodin, Lalitha Dabbiru

Proceedings Volume 11401, 1140104 (2020) https://doi.org/10.1117/12.2557928

KEYWORDS: Data modeling, Image segmentation, Image compression, Performance modeling, Data compression, Data storage, Image processing algorithms and systems, Image quality, Micro unmanned aerial vehicles

Read Abstract +

Proceedings Article | 2 May 2019 Presentation + Paper

Multi-LiDAR placement, calibration, co-registration, and processing on a Subaru Forester for off-road autonomous vehicles operations

Will Meadows, Christopher Hudson, Chris Goodin, Lalitha Dabbiru, Brandon Powell, Matt Doude, Daniel Carruth, Muhammad Islam, John Ball, Bo Tang

Proceedings Volume 11009, 110090J (2019) https://doi.org/10.1117/12.2518915

KEYWORDS: LIDAR, Sensors, Calibration, Neural networks, Data processing, Environmental sensing, Optical simulations

Read Abstract +

Showing 5 of 8 publications

View contact details

UPDATE YOUR PROFILE

Is this your profile? Update it now.

Sign into your SPIE.org account

Don’t have a profile and want one?

Create an account on SPIE.org

Keywords/Phrases

Search In:

Publication Years