This paper presents a comprehensive review of ground agricultural robotic systems and applications with special focus on harvesting that span research and commercial products and results, as well as their enabling technologies. The majority of literature concerns the development of crop detection, field navigation via vision and their related challenges. Health monitoring, yield estimation, water status inspection, seed planting and weed removal are frequently encountered tasks. Regarding robotic harvesting, apples, strawberries, tomatoes and sweet peppers are mainly the crops considered in publications, research projects and commercial products. The reported harvesting agricultural robotic solutions, typically consist of a mobile platform, a single robotic arm/manipulator and various navigation/vision systems. This paper reviews reported development of specific functionalities and hardware, typically required by an operating agricultural robot harvester; they include (a) vision systems, (b) motion planning/navigation methodologies (for the robotic platform and/or arm), (c) Human-Robot-Interaction (HRI) strategies with 3D visualization, (d) system operation planning & grasping strategies and (e) robotic end-effector/gripper design. Clearly, automated agriculture and specifically autonomous harvesting via robotic systems is a research area that remains wide open, offering several challenges where new contributions can be made.

Automation of vineyards cultivation necessitates for mobile robots to retain accurate localization system. The paper introduces a stereo vision-based Graph-Simultaneous Localization and Mapping (Graph-SLAM) pipeline custom-tailored to the specificities of vineyard fields. Graph-SLAM is reinforced with a Loop Closure Detection (LCD) based on semantic segmentation of the vine trees. The Mask R-CNN network is applied to segment the trunk regions of images, on which unique visual features are extracted. These features are used to populate the bag of visual words (BoVW s) retained on the formulated graph. A nearest neighbor search is applied to each query trunk-image to associate each unique feature descriptor with the corresponding node in the graph using a voting procedure. We apply a probabilistic method to select the most suitable loop closing pair and, upon an LCD appearance, the 3D points of the trunks are employed to estimate the loop closure constraint to the graph. The traceable features on trunk segments drastically reduce the number of retained BoVWs, which in turn expedites significantly the loop closure and graph optimization, rendering our method suitable for large scale mapping in vineyards. The pipeline has been evaluated on several data sequences gathered from real vineyards, in different seasons, when the appearance of vine trees vary significantly, and exhibited robust mapping in long distances.

This paper presents an innovative approach to optimize the trajectories of a robotic surveillance system, employing three different optimization methods: genetic algorithm (GA), particle swarm optimization (PSO), and pattern search (PS). The research addresses the challenge of efficiently planning routes for a LiDAR-equipped mobile robot to effectively cover target areas taking into account the capabilities and limitations of sensors and robots. The findings demonstrate the effectiveness of these trajectory optimization approaches, significantly improving detection efficiency and coverage of critical areas. Furthermore, it is observed that, among the three techniques, pattern search quickly obtains feasible solutions in environments with good initial trajectories. On the contrary, in cases where the initial trajectory is suboptimal or the environment is complex, PSO works better. For example, in the high complexity map evaluated, PSO achieves 86.7% spatial coverage, compared to 85% and 84% for PS and GA, respectively. On low- and medium-complexity maps, PS is 15.7 and 18 s faster in trajectory optimization than the second fastest algorithm, which is PSO in both cases. Furthermore, the fitness function of this proposal has been compared with that of previous works, obtaining better results.

Driven by the increasing food demand and the need for higher-quality cultivation, precision agriculture grows steadily during the last decade. It involves the application of mobile robots and intelligent robotic technologies in various agricultural field tasks, concerning a variety of crop types. Aiming at compensating for the lack of selective robotic harvesting solutions regarding the high-value crop of grapes, the EU-funded project BACCHUS ¹ develops an intelligent mobile robotic system, comprising two independent and cooperative robots: one for the grape inspection and collection of valuable data regarding their maturity level, and one for the bimanual harvesting of grapes in a human-inspired manner. Validated via real-field trials, the proposed autonomous system pushes forward the precision agriculture application for a particularly sensitive crop type in the challenging and heavily cluttered environment of vineyards, facilitating the selective harvesting of high-quality grapes.

In this letter, we present a novel method for key-frame extraction in assembly processes, with the support of visual sensors. The method is based on the construction of a sequence of weighted graphs containing the semantics of the underlying scene, encoded as primitive temporal and spatial relations between the involved objects. The objects are modeled as three-dimensional ellipsoids, with the formulation of the minimum volume covering ellipsoid problem as a convex optimization program, which is solved by an efficient algorithm. The new models are used in the sequence of actions-relations to increase the accuracy of the key-frame extraction, by extending the states encoded in the graphs. We present and analyze the method, followed by preliminary experimental results, and provide significant evidence on its potential.