# Probabilistic Sampling Networks for Hybrid Structure Planning in Semi-Structured Environments

## TL;DR

A new robot motion planning method combines a probabilistic sampling network and an enhanced potential field to improve performance in complex environments.

## Contribution

A hybrid motion planner using PSNet and EAPF is proposed, outperforming existing methods in high-dimensional planning tasks.

## Key findings

- The method achieves higher success rates and stable collision avoidance in 2-D and 6-DOF tasks.
- It improves adaptability and robustness for industrial robots in dynamic manufacturing settings.
- The hybrid approach enables efficient path planning and obstacle avoidance in complex scenarios.

## Abstract

What are the main findings?
A hybrid structure planning method based on a Probabilistic Sampling Network (PSNet) and an Enhanced Artificial Potential Field (EAPF) is proposed to address high-dimensional robot motion planning in semi-structured environments.Experiments demonstrate that the proposed method outperforms MPNet and RRT-Connect in both 2-D point-mass robot and 6-DOF manipulator tasks, achieving higher success rates and more stable collision avoidance.

A hybrid structure planning method based on a Probabilistic Sampling Network (PSNet) and an Enhanced Artificial Potential Field (EAPF) is proposed to address high-dimensional robot motion planning in semi-structured environments.

Experiments demonstrate that the proposed method outperforms MPNet and RRT-Connect in both 2-D point-mass robot and 6-DOF manipulator tasks, achieving higher success rates and more stable collision avoidance.

What is the implication of the main finding?
The proposed approach enhances the adaptability and robustness of industrial robots in intelligent manufacturing, maintaining efficient path planning in dynamic and complex scenarios.This study provides a new perspective for integrating learning-based methods with classical planning techniques, laying the foundation for future applications in autonomous robotic operations and human–robot collaboration.

The proposed approach enhances the adaptability and robustness of industrial robots in intelligent manufacturing, maintaining efficient path planning in dynamic and complex scenarios.

This study provides a new perspective for integrating learning-based methods with classical planning techniques, laying the foundation for future applications in autonomous robotic operations and human–robot collaboration.

The advancement of adaptable industrial robots in intelligent manufacturing is hindered by the inefficiency of traditional motion planning methods in high-dimensional spaces. Therefore, a Dempster–Shafer evidence theory-based hybrid motion planner is proposed, in which a probabilistic sampling network (PSNet) and an enhanced artificial potential field (EAPF) cooperate with each other to improve the planning performance. The PSNet architecture comprises two modules: a motion planning module (MPM) and a fusion sampling module (FSM). The MPM utilizes sensor data alongside the robot’s current and target configurations to recursively generate diverse multimodal distributions of the next configuration. Based on the distribution information, the FSM was used as a decision-maker to ultimately generate globally connectable paths. Moreover, the FSM is equipped to correct collision path points caused by network inaccuracies through Gaussian resampling. Simultaneously, an augmented artificial potential field with a dynamic rotational field is deployed to repair local paths when worst-case collision scenarios occur. This collaborative strategy harmoniously unites the complementary strengths of both components, thereby enhancing the overall resilience and adaptability of the motion planning system. Experiments were conducted in various environments. The results demonstrate that the proposed method can quickly find directly connectable paths in diverse environments while reliably avoiding sudden obstacles.