Multi-log grasping using reinforcement learning and virtual visual servoing

TL;DR

This paper presents a reinforcement learning and virtual visual servoing approach for multi-log grasping in simulated environments, achieving high success rates in complex outdoor-like scenarios.

Contribution

It introduces a novel method combining virtual visual servoing with reinforcement learning for multi-log grasping, simplifying domain transfer and reducing computational demands.

Findings

95% success rate in multi-log grasping tasks

Effective separation of image segmentation from crane control

Use of virtual camera and 3D reconstruction for robust control

Abstract

We explore multi-log grasping using reinforcement learning and virtual visual servoing for automated forwarding in a simulated environment. Automation of forest processes is a major challenge, and many techniques regarding robot control pose different challenges due to the unstructured and harsh outdoor environment. Grasping multiple logs involves various problems of dynamics and path planning, where understanding the interaction between the grapple, logs, terrain, and obstacles requires visual information. To address these challenges, we separate image segmentation from crane control and utilise a virtual camera to provide an image stream from reconstructed 3D data. We use Cartesian control to simplify domain transfer to real-world applications. Since log piles are static, visual servoing using a 3D reconstruction of the pile and its surroundings is equivalent to using real camera data until the point of grasping. This relaxes the limits on computational resources and time for the challenge of image segmentation and allows for collecting data in situations where the log piles are not occluded. The disadvantage is the lack of information during grasping. We demonstrate that this problem is manageable and present an agent that is 95% successful in picking one or several logs from challenging piles of 2--5 logs.