Path Generation for Wheeled Robots Autonomous Navigation on Vegetated Terrain

Zhuozhu Jian,Zejia Liu,Haoyu Shao,Xueqian Wang,Xinlei Chen,Bin Liang

2023-11-29

Abstract:Wheeled robot navigation has been widely used in urban environments, but little research has been conducted on its navigation in wild vegetation. External sensors (LiDAR, camera etc.) are often used to construct point cloud map of the surrounding environment, however, the supporting rigid ground used for travelling cannot be detected due to the occlusion of vegetation. This often causes unsafe or not smooth path during planning process. To address the drawback, we propose the PE-RRT* algorithm, which effectively combines a novel support plane estimation method and sampling algorithm to generate real-time feasible and safe path in vegetation environments. In order to accurately estimate the support plane, we combine external perception and proprioception, and use Multivariate Gaussian Processe Regression (MV-GPR) to estimate the terrain at the sampling nodes. We build a physical experimental platform and conduct experiments in different outdoor environments. Experimental results show that our method has high safety, robustness and generalization.

Robotics

What problem does this paper attempt to address?

This paper attempts to address the challenges encountered in autonomous navigation of wheeled robots on vegetation-covered terrain. Specifically, due to the occlusion by vegetation, existing external sensors (such as LiDAR, cameras, etc.) cannot accurately detect the supporting ground, leading to unsafe or unsmooth paths generated during the path planning process. To solve this problem, the authors propose the PE-RRT* algorithm, which combines a new supporting plane estimation method and a sampling algorithm, capable of generating feasible and safe paths in real-time in vegetated environments.

### Main Issues:
1. **Vegetation Occlusion Problem**: In vegetation-covered environments, external sensors cannot accurately detect the supporting ground, leading to unsafe or unsmooth paths generated during the path planning process.
2. **Limitations of Existing Methods**: Existing path planning methods usually treat vegetation as obstacles, which is overly conservative because wheeled robots can pass through shorter, penetrable vegetation. Additionally, traditional methods often require the construction of a priori traversable maps, which is time-consuming, especially when accurate estimation of the supporting ground is needed.

### Solutions:
1. **PE-RRT* Algorithm**: This algorithm combines a new supporting plane estimation method and a sampling algorithm, capable of generating feasible and safe paths in real-time in vegetated environments.
2. **Supporting Plane Estimation**: By fusing proprioception and exteroception, and using Multivariate Gaussian Process Regression (MV-GPR) to estimate the terrain at the sampling nodes.
3. **Enhanced Safety**: Introducing a safety inflation radius in the sampling algorithm to enhance the safety of the generated paths.

### Experimental Validation:
The authors built a physical experimental platform and conducted experiments in different outdoor environments. The experimental results show that this method has high safety, robustness, and generalization ability.