Papers
Topics
Authors
Recent
Search
2000 character limit reached

Advancement and Field Evaluation of a Dual-arm Apple Harvesting Robot

Published 6 Jun 2025 in cs.RO | (2506.05714v1)

Abstract: Apples are among the most widely consumed fruits worldwide. Currently, apple harvesting fully relies on manual labor, which is costly, drudging, and hazardous to workers. Hence, robotic harvesting has attracted increasing attention in recent years. However, existing systems still fall short in terms of performance, effectiveness, and reliability for complex orchard environments. In this work, we present the development and evaluation of a dual-arm harvesting robot. The system integrates a ToF camera, two 4DOF robotic arms, a centralized vacuum system, and a post-harvest handling module. During harvesting, suction force is dynamically assigned to either arm via the vacuum system, enabling efficient apple detachment while reducing power consumption and noise. Compared to our previous design, we incorporated a platform movement mechanism that enables both in-out and up-down adjustments, enhancing the robot's dexterity and adaptability to varying canopy structures. On the algorithmic side, we developed a robust apple localization pipeline that combines a foundation-model-based detector, segmentation, and clustering-based depth estimation, which improves performance in orchards. Additionally, pressure sensors were integrated into the system, and a novel dual-arm coordination strategy was introduced to respond to harvest failures based on sensor feedback, further improving picking efficiency. Field demos were conducted in two commercial orchards in MI, USA, with different canopy structures. The system achieved success rates of 0.807 and 0.797, with an average picking cycle time of 5.97s. The proposed strategy reduced harvest time by 28% compared to a single-arm baseline. The dual-arm harvesting robot enhances the reliability and efficiency of apple picking. With further advancements, the system holds strong potential for autonomous operation and commercialization for the apple industry.

Summary

  • The paper introduces a dual-arm configuration that reduces harvest time by 28% compared to a single-arm system.
  • The paper details an innovative integration of ToF camera, sensor feedback, and a centralized vacuum system for efficient apple localization and detachment.
  • Field evaluations in Michigan orchards showed a success rate of ~80% and an average picking cycle time of 5.97 seconds, underscoring its commercial viability.

Development and Evaluation of a Dual-arm Apple Harvesting Robot

The manuscript presents a comprehensive study on the development and field evaluation of an advanced dual-arm apple harvesting robot. This research addresses the persistent challenge of automating apple harvesting in commercial orchards, particularly focusing on enhancing the reliability, cost-effectiveness, and adaptability of robotic systems in complex outdoor environments. The proposed system integrates state-of-the-art technology, including a Time-of-Flight (ToF) camera, two robotic arms each with four degrees of freedom, and a centralized vacuum system that dynamically assigns suction force for efficient fruit detachment.

System Design and Innovations

Key innovations in the system design include the introduction of a dual-arm configuration, which significantly enhances harvesting efficiency by allowing parallel operations. The robot's platform movement mechanism allows in-out and up-down adjustments, which increases its adaptability to diverse canopy structures. The dual-arm system coordinates the operations of both arms using a centralized vacuum source and relies on pressure sensor feedback to adjust to harvesting failures, providing operational independence and maximizing efficiency.

On the algorithmic front, the paper describes the implementation of a robust apple localization pipeline. This pipeline leverages a foundation-model-based apple detector, pixel-wise segmentation, and clustering-based depth estimation to improve localization accuracy, particularly in challenging outdoor conditions. This approach addresses common issues such as canopy occlusion and extreme lighting, enhancing the system's robustness and reliability.

Field Evaluation Results

Field tests were conducted in two commercial orchards in Michigan, using different canopy structures. The system achieved success rates of 80.7% and 79.7%, with an average picking cycle time of 5.97 seconds. These results provide empirical evidence of the system's capability to perform efficient and speedy apple harvesting under real-world commercial conditions. The dual-arm coordination strategy demonstrated a reduction in harvest time by 28% compared to a single-arm baseline, showcasing the potential for enhanced efficiency in future autonomous deployments.

Implications and Future Directions

This research advances the field of agricultural robotics by addressing the primary challenges faced in robotic fruit harvesting: perception robustness and operational efficiency. The dual-arm configuration offers a scalable solution that can adapt to various orchard structures, while the robust perception system ensures high accuracy in fruit localization. The evaluation of this system in different orchard environments suggests promising potential for commercial deployment.

Future developments could explore further optimization of the perception algorithms under high-occlusion scenarios and refine the hardware design to minimize canopy interference. Additionally, integrating the robot with autonomous navigation systems could enable continuous, seamless harvesting, further increasing operational efficiency. These enhancements would support the transition to fully autonomous apple harvesting systems, potentially reducing the economic and labor-related challenges faced by the apple industry.

Statistically, the results reaffirm the feasibility of using robotic systems to replace manual labor in apple harvesting, presenting a viable solution to rising labor costs and labor shortages. In conjunction with continued advancements in perception and control algorithms, these robotic systems hold the promise of revolutionizing the apple harvesting process, improving both yield efficiency and fruit quality.

In summary, this paper contributes significant knowledge to the field of robotic automation in agriculture, providing a well-documented case study of a dual-arm harvesting robot and setting a foundation for future research and development in this area.

Paper to Video (Beta)

No one has generated a video about this paper yet.

Whiteboard

No one has generated a whiteboard explanation for this paper yet.

Open Problems

We haven't generated a list of open problems mentioned in this paper yet.

Collections

Sign up for free to add this paper to one or more collections.