Path Planning Based Navigation Using LIDAR for an Ackerman Unmanned Ground Vehicle

Aguilar, Wilbert G.; Sandoval, Sebastián; Limaico, Alex; Villegas-Pico, Martin; Asimbaya, Israel

doi:10.1007/978-3-030-27541-9_33

Cited by 10 publications

(8 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thresholding and contours techniques are used to extract the 2D bounding box coordinates in the birds-eye-view image frame. Moreover, the obtained 2D coordinates have been used in the back projection module (BPM) to extract the 3D obstacle information and display it as a 3D bounding volume using Equations ( 9)- (11). Qualitative results of the proposed model are shown in the figures below where Figure 9 shows 3D object detection results from the KITTI dataset and Figure 10 shows 3D object detection results from the Ouster Lidar-64 dataset.…”

Section: Object Detection Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Transfer Learning Based Semantic Segmentation for 3D Object Detection from Point Cloud

Imad

Doukhi

Lee

2021

Sensors

View full text Add to dashboard Cite

Three-dimensional object detection utilizing LiDAR point cloud data is an indispensable part of autonomous driving perception systems. Point cloud-based 3D object detection has been a better replacement for higher accuracy than cameras during nighttime. However, most LiDAR-based 3D object methods work in a supervised manner, which means their state-of-the-art performance relies heavily on a large-scale and well-labeled dataset, while these annotated datasets could be expensive to obtain and only accessible in the limited scenario. Transfer learning is a promising approach to reduce the large-scale training datasets requirement, but existing transfer learning object detectors are primarily for 2D object detection rather than 3D. In this work, we utilize the 3D point cloud data more effectively by representing the birds-eye-view (BEV) scene and propose a transfer learning based point cloud semantic segmentation for 3D object detection. The proposed model minimizes the need for large-scale training datasets and consequently reduces the training time. First, a preprocessing stage filters the raw point cloud data to a BEV map within a specific field of view. Second, the transfer learning stage uses knowledge from the previously learned classification task (with more data for training) and generalizes the semantic segmentation-based 2D object detection task. Finally, 2D detection results from the BEV image have been back-projected into 3D in the postprocessing stage. We verify results on two datasets: the KITTI 3D object detection dataset and the Ouster LiDAR-64 dataset, thus demonstrating that the proposed method is highly competitive in terms of mean average precision (mAP up to 70%) while still running at more than 30 frames per second (FPS).

show abstract

Section: Object Detection Resultsmentioning

confidence: 99%

“…Existing studies on transfer learning mainly focused on 2D [9,10] object detection. Nevertheless, 2D object detection does not provide depth information, a requisite for autonomous driving tasks such as path planning [11] and collision avoidance [12].…”

Section: Introductionmentioning

confidence: 99%

Transfer Learning Based Semantic Segmentation for 3D Object Detection from Point Cloud

Imad

Doukhi

Lee

2021

Sensors

View full text Add to dashboard Cite

show abstract

“…The work by Chen et al 23 evaluates a system of planning a route concurrently for UAV and UGV, with the limitations to the space of road networks. The works by Aguilar et al and Lakas et al 24,25 focus on algorithms of route planning in moving UGV, navigation, and avoidance of obstacles using laser radar or visual sensors. And last but not least, a comprehensive planning system for maneuvers of structured military forces and equipment and the interconnection of maneuvers of units and their fire support are given in the work by Harder.…”

Section: Literature Reviewmentioning

confidence: 99%

Multiple maneuver model of cooperating ground combat troops

Nohel

Stodola

Flasar

et al. 2022

Journal of Defense Modeling & Simulation

View full text Add to dashboard Cite

The article describes the possibilities of using the Maneuver Control System CZ (MCS CZ) in creating a multiple maneuver of a group of cooperating units. It is a tactical information system designed to support the decision-making of commanders in order to effectively deploy their units on the battlefield. The system evaluates the geographical and hydrometeorological battlefield influences as well as the enemy’s situation to calculate the maneuver axes of a unit composed of up to three teams with the same goal. To verify the calculations of mathematical-algorithmic models of the system, two measurements were performed on model situations. In the first of them, a group of three unmanned systems carried out an attack on an identified delivery system of weapons of mass destruction. The second model situation included a hasty attack by three infantry squads on anti-aircraft equipment. In both cases, the resulting group of maneuvers used the fastest and safe axes of advance, spatially synchronized according to tactical principles. In both cases, the enemy’s weapon systems were quickly attacked and destroyed immediately after their identification from three directions. The commander’s decision-making process was largely replaced by the MCS CZ calculations in model situations, which significantly shortened its duration.

show abstract

“…The optimization algorithms for pathfinding for ground robotics [20][21][22][23][24], aerial vehicles [25][26][27], and underwater vehicles [28,29] includes a wide range of applications. The most well-known applications for autonomous vehicles are obstacle avoidance, path planning, localization, navigation, sensing, and communication, which works on preessential maps related to the environment; they also play a vital role in communication relay, aviation industry for surveillance, and loitering dominated missions.…”

Section: Scholarly Contributions and Applicationsmentioning

confidence: 99%

A Consolidated Review of Path Planning and Optimization Techniques: Technical Perspectives and Future Directions

et al. 2021

View full text Add to dashboard Cite

In this paper, a review on the three most important communication techniques (ground, aerial, and underwater vehicles) has been presented that throws light on trajectory planning, its optimization, and various issues in a summarized way. This kind of extensive research is not often seen in the literature, so an effort has been made for readers interested in path planning to fill the gap. Moreover, optimization techniques suitable for implementing ground, aerial, and underwater vehicles are also a part of this review. This paper covers the numerical, bio-inspired techniques and their hybridization with each other for each of the dimensions mentioned. The paper provides a consolidated platform, where plenty of available research on-ground autonomous vehicle and their trajectory optimization with the extension for aerial and underwater vehicles are documented.

show abstract

Path Planning Based Navigation Using LIDAR for an Ackerman Unmanned Ground Vehicle

Cited by 10 publications

References 15 publications

Transfer Learning Based Semantic Segmentation for 3D Object Detection from Point Cloud

Transfer Learning Based Semantic Segmentation for 3D Object Detection from Point Cloud

Multiple maneuver model of cooperating ground combat troops

A Consolidated Review of Path Planning and Optimization Techniques: Technical Perspectives and Future Directions

Contact Info

Product

Resources

About