3D Navigation Mesh Generation for Path Planning in Uneven Terrain

Pütz, Sebastian; Wiemann, Thomas; Sprickerhof, Jochen; Hertzberg, Joachim

doi:10.1016/j.ifacol.2016.07.734

Cited by 40 publications

(15 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A few traditional vision works have used methods such as Markov Random fields [16] on fused data from 3D lidar and RGB data and triangular mesh reconstruction [17] from 3D point clouds to analyze surface roughness and traversability. All previous learning-based works in computer vision for traversability prediction fall into a combination of supervised/unsupervised and classification/regression categories (discussed below).…”

Section: A Characterizing Traversabilitymentioning

confidence: 99%

TerraPN: Unstructured Terrain Navigation using Online Self-Supervised Learning

Sathyamoorthy¹,

Weerakoon²,

Guan³

et al. 2022

Preprint

View full text Add to dashboard Cite

We present TerraPN, a novel method to learn the surface characteristics (texture, bumpiness, deformability, etc.) of complex outdoor terrains for autonomous robot navigation. Our method predicts navigability cost maps for different surfaces using patches of RGB images, odometry, and IMU data. Our method dynamically varies the resolution of the output cost map based on the scene to improve its computational efficiency. We present a novel extension to the Dynamic-Window Approach (DWA-O) to account for a surface's navigability cost while computing robot trajectories. DWA-O also dynamically modulates the robot's acceleration limits based on the variation in the robot-terrain interactions. In terms of perception, our method learns to predict navigability costs in ∼ 20 minutes for five different surfaces, compared to 3-4 hours for previous scene segmentation methods, and leads to a decrease in inference time. In terms of navigation, our method outperforms previous works in terms of vibration costs and generates robot velocities suitable for different surfaces.• A novel method that trains a neural network in an online self-supervised manner to learn a terrain's surface

show abstract

Section: A Characterizing Traversabilitymentioning

confidence: 99%

TerraPN: Unstructured Terrain Navigation using Online Self-Supervised Learning

Sathyamoorthy¹,

Weerakoon²,

Guan³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Gu et al [18] suggested to utilizing a 2.5D map to assess the traversability of different terrains. Differently, in [5], a 3D mesh map was ready for online path planning in rough terrain by a map generation method based on 3D point cloud. More recently, Sock et al [19] proposed a traversable map that simply distinguished different terrains in a probabilistic manner.…”

Section: A Environment Representationmentioning

confidence: 99%

“…For the wheelchair robot navigation, one fundamental and critical part is a suitable representation of the complex environment. Recently, a vast amount of research has been devoted to developing various maps [4], [5]. Different maps have their own merits for different applications [6], [7].…”

Section: Introductionmentioning

confidence: 99%

Stable Autonomous Robotic Wheelchair Navigation in the Environment With Slope Way

Wang

Xia

Meng

2020

IEEE Trans. Veh. Technol.

View full text Add to dashboard Cite

In this article, we present a path planning approach that is capable of generating a feasible trajectory for stable robotic wheelchair navigation in the environment with slope way. Firstly, the environment is modeled by a lightweight navigation map, with which the proposed sampling-based path planning scheme with a modified extension function can generate a feasible path. Then, the path is further optimized by the proposed utility function involving the human comfort and the path cost. To improve the searching efficiency of an optimal trajectory, we present an adaptive weighting Gaussian Mixture Model (GMM) based sampling strategy. Particularly, the weights of the components in GMM are adjusted adaptively in the planning process. It is also worth noting that the proposed sampling-based planning paradigm can indicate the unsafe regions in the navigation map, which forms a traversable map and further guarantees the safety of the wheelchair robot navigation. Furthermore, the effectiveness and the efficiency of the proposed path planning method are verified in both simulation and real-world experiments.

show abstract

“…In order to displace a hardware complex along a working, among various options, caterpillar and pneumo-wheeled platforms, the most widely used are those four-wheeled and six-wheeled [12]. Application of pneumo-wheeled platforms minimizes energy costs when moving the equipment.…”

Section: Literature Review and Problem Statementmentioning

confidence: 99%