Complex Terrain Navigation via Model Error Prediction

Polevoy, Adam; Knuth, Craig; Popek, Katie M.; Katyal, Kapil D.

doi:10.1109/icra46639.2022.9811644

Cited by 11 publications

(4 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Several 2.5D maps include additional terrain information. Waibel et al (2022) explores terrain roughness through point clouds, while Polevoy et al (2022) predicts nonrigid terrain traversability by learning trajectory model errors. Gasparino et al (2022) directly learns the traversability of the environment to generate traversable paths, and Dang et al (2020) evaluates the performance of each off‐road path to assess its traversability for vehicles.…”

Section: Related Workmentioning

confidence: 99%

Motion planning for off‐road autonomous driving based on human‐like cognition and weight adaptation

Wang,

Gong,

Gong

et al. 2024

Journal of Field Robotics

View full text Add to dashboard Cite

Driving in an off‐road environment is challenging for autonomous vehicles due to the complex and varied terrain. To ensure stable and efficient travel, the vehicle requires consideration and balancing of environmental factors, such as undulations, roughness, and obstacles, to generate optimal trajectories that can adapt to changing scenarios. However, traditional motion planners often utilize a fixed cost function for trajectory optimization, making it difficult to adapt to different driving strategies in challenging irregular terrains and uncommon scenarios. To address these issues, we propose an adaptive motion planner based on human‐like cognition and cost evaluation for off‐road driving. First, we construct a multilayer map describing different features of off‐road terrains, including terrain elevation, roughness, obstacle, and artificial potential field map. Subsequently, we employ a convolutional neural network‐long short‐term memory network to learn the trajectories planned by human drivers in various off‐road scenarios. Then, based on human‐like generated trajectories in different environments, we design a primitive‐based trajectory planner that aims to mimic human trajectories and cost weight selection, generating trajectories that are consistent with the dynamics of off‐road vehicles. Finally, we compute optimal cost weights and select and extend behavioral primitives to generate highly adaptive, stable, and efficient trajectories. We validate the effectiveness of the proposed method through experiments in a desert off‐road environment with complex terrain and varying road conditions. The experimental results show that the proposed human‐like motion planner has excellent adaptability to different off‐road conditions. It shows real‐time operation, greater stability, and more human‐like planning ability in diverse and challenging scenarios.

show abstract

Section: Related Workmentioning

confidence: 99%

Motion planning for off‐road autonomous driving based on human‐like cognition and weight adaptation

Wang,

Gong,

Gong

et al. 2024

Journal of Field Robotics

View full text Add to dashboard Cite

show abstract

“…[10] assumes that ground heights smoothly vary and terrain classes tend to cluster, and uses Markov random fields to infer the supporting ground surface for navigation based on LiDAR points. [11] defines a regression problem which estimates predicted error between the realized odometry readings and the predicted trajectory. And they utilize machine learning techniques to predict model error associated with an RGB image.…”

Section: Related Workmentioning

confidence: 99%

Parametric Path Optimization for Wheeled Robots Navigation

Jian

Zhang

et al. 2022

2022 International Conference on Robotics and Automation (ICRA)

View full text Add to dashboard Cite

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. The source code is released for the reference of the community 1 .

show abstract

“…This study only considers static workplace obstructions; moving obstructions caused by moving objects are not included. This study only considered unexpectedly appearing static obstacles, although model-based predictive controller (MBPC) using neural networks and ultrasonic sensors is also utilized to guide mobile robots around unexpectedly appearing static obstacles in their environment [62][63][64][65][66][67][68][69][70][71][72][73][74][75][76][77]. The Dynamic Artificial Neural Network (DANN) approach is used for motion planning for mobile robot paths through [78][79][80].…”

Section: Introductionmentioning

confidence: 99%

Simulation of Obstacle Avoidance Robots

Ibrahim Bello,

Baballe

2023

J Mat Sci Eng Technol

View full text Add to dashboard Cite

The first function of the system is to detect the presence of obstacles. When the user activates the system using the power ON/OFF switch, the Arduino microcontroller will read the data. When the ultrasonic sensor detects the presence of an obstacle in the process of moving forward, the robot will move backward. If the robot does not sense any obstacle, that is, if the distance between an obstacle and it is wide, it will then move forward again until it senses an obstacle before it stops. C programming is used for Arduino board applications to develop the program for the whole system's operation. There are three light-emitting diodes; the first one shows the amount of charge in the batteries, while the remaining two show that if the robot is moving forward, one of the two LEDs will be on, and if it is moving backward, one of the LEDs will also be on. There is also a power source unit that is used to charge the batteries used in the system.

show abstract

Complex Terrain Navigation via Model Error Prediction

Cited by 11 publications

References 19 publications

Motion planning for off‐road autonomous driving based on human‐like cognition and weight adaptation

Motion planning for off‐road autonomous driving based on human‐like cognition and weight adaptation

Parametric Path Optimization for Wheeled Robots Navigation

Simulation of Obstacle Avoidance Robots

Contact Info

Product

Resources

About