Light-Weight Pointcloud Representation with Sparse Gaussian Process

Ali, Mahmoud N.; Liu, Lantao

doi:10.1109/icra48891.2023.10161111

Cited by 7 publications

(2 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our proposed GP-subgoal recommendation policy relies on our earlier work presented in [23], [24], where we transformed pointcloud data into an occupancy surface and modeled it using a Sparse Gaussian Process (SGP). Within this approach, the occupancy surface takes the form of a 2D circular surface centered around the sensor origin and has a predefined radius of r oc .…”

Section: A Sgp Occupancy Surface Representationmentioning

confidence: 99%

“…2, we present a concrete example of the SGP occupancy model applied to our Jackal robot, which is equipped with a Velodyne VLP-16 LiDAR and located in an unknown cluttered environment, as depicted in (10). The precision of the SGP occupancy model is intensively evaluated in our previous work [23], where the results showed that an SGP occupancy model comprising of 400 inducing points generates a reconstructed point cloud with an average error of approximately 12 cm.…”

Section: A Sgp Occupancy Surface Representationmentioning

confidence: 99%

See 1 more Smart Citation

A Robust and Fast Occlusion-based Frontier Method for Autonomous Navigation in Unknown Cluttered Environments

Mohammad

Bezzo

2022

2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

View full text Add to dashboard Cite

Robotic navigation in unknown, cluttered environments with limited sensing capabilities poses significant challenges in robotics. Local trajectory optimization methods, such as Model Predictive Path Intergal (MPPI), are a promising solution to this challenge. However, global guidance is required to ensure effective navigation, especially when encountering challenging environmental conditions or navigating beyond the planning horizon. This study presents the GP-MPPI, an online learning-based control strategy that integrates MPPI with a local perception model based on Sparse Gaussian Process (SGP).The key idea is to leverage the learning capability of SGP to construct a variance (uncertainty) surface, which enables the robot to learn about the navigable space surrounding it, identify a set of suggested subgoals, and ultimately recommend the optimal subgoal that minimizes a predefined cost function to the local MPPI planner. Afterward, MPPI computes the optimal control sequence that satisfies the robot and collision avoidance constraints. Such an approach eliminates the necessity of a global map of the environment or an offline training process. We validate the efficiency and robustness of our proposed control strategy through both simulated and real-world experiments of 2D autonomous navigation tasks in complex unknown environments, demonstrating its superiority in guiding the robot safely towards its desired goal while avoiding obstacles and escaping entrapment in local minima. The GPU implementation of GP-MPPI, including the supplementary video, is available at https://github.com/IhabMohamed/GP-MPPI.

show abstract

Section: A Sgp Occupancy Surface Representationmentioning

confidence: 99%

Section: A Sgp Occupancy Surface Representationmentioning

confidence: 99%

A Robust and Fast Occlusion-based Frontier Method for Autonomous Navigation in Unknown Cluttered Environments

Mohammad

Bezzo

2022

2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

View full text Add to dashboard Cite

show abstract

Improving Brain Tumor Classification: An Approach Integrating Pre-Trained CNN Models and Machine Learning Algorithms

Shoaib,

Zhao,

Emara

et al. 2024

Heliyon

View full text Add to dashboard Cite

GP-Frontier for Local Mapless Navigation

Ali

Liu

2023

2023 IEEE International Conference on Robotics and Automation (ICRA)

Self Cite

View full text Add to dashboard Cite

We propose a new method for autonomous navigation in uneven terrains by utilizing a sparse Gaussian Process (SGP) based local perception model. The SGP local perception model is trained on local ranging observation (pointcloud) to learn the terrain elevation profile and extract the feasible navigation subgoals around the robot. Subsequently, a cost function, which prioritizes the safety of the robot in terms of keeping the robot's roll and pitch angles bounded within a specified range, is used to select a safety-aware subgoal that leads the robot to its final destination. The algorithm is designed to run in real-time and is intensively evaluated in simulation and realworld experiments. The results compellingly demonstrate that our proposed algorithm consistently navigates uneven terrains with high efficiency and surpasses the performance of other planners. The code can be found here: https://rb.gy/3ov2r8.

show abstract

Light-Weight Pointcloud Representation with Sparse Gaussian Process

Cited by 7 publications

References 17 publications

A Robust and Fast Occlusion-based Frontier Method for Autonomous Navigation in Unknown Cluttered Environments

A Robust and Fast Occlusion-based Frontier Method for Autonomous Navigation in Unknown Cluttered Environments

Improving Brain Tumor Classification: An Approach Integrating Pre-Trained CNN Models and Machine Learning Algorithms

GP-Frontier for Local Mapless Navigation

Contact Info

Product

Resources

About