The characterization of the detection capability assumes significance when the reliable monitoring of the region of interest by a non-contact sensor is a safety-relevant function. This paper introduces new validation scores that evaluate the detection capability of non-contact sensors intended to be applied to outdoor machines. The scores quantify, in terms of safety, the suitability of the sensor for the intended implementation in an environmental perception system of (highly) automated machines. This was achieved by developing an extension to the new Real Environment Detection Area (REDA) method and linking the methodology with the sensor standard IEC/TS 62998-1. The extension includes point-by-point and statistic-based error evaluation which leads to the Usability-Score, Availability-Score, and Reliability-Score. By applying the principle in the agricultural sector using ISO 18497 and linking this with data from a real outdoor test stand, it was possible to show that the validation scores offer a generic approach to quantify the detection capability and express this in a machine manufacturer-oriented manner. The findings of this study have significant implications for the advancement of safety-related sensor systems integrated into machines operating in complex environments. In order to achieve full implementation, it is necessary to define in the standards which score is required for each Performance Level (PL).

INHALT Electrical Agricultural Machines Structuring of electrified agricultural machine systems – Diversity of solutions and analysis methods .....1 GridCON2 – Development of a Cable Drum Vehicle Concept to Power 1MW Fully Electric Agricultural Swarms ..... 11 GridCON Swarm – Development of a Grid Connected Fully Autonomous Agricultural Production System ..... 17 Fully electric Tractor with 1000 kWh battery capacity ..... 23 Soil and Modelling The Integration of a Scientific Soil Compaction Risk Indicator (TERRANIMO) into a Holistic Tractor and Implement Optimization System (CEMOS) .....29 Identification of draft force characteristics for a tillage tine with variable geometry ..... 37 Calibration of soil models within the Discrete Element Method (DEM) ..... 45 Automation and Optimization of Working Speed and Depth in Agricultural Soil Tillage with a Model Predictive Control based on Machine Learning ..... 55 Synchronising machine adjustments of combine harvesters for higher fleet performance ..... 65 A generic approach to bridge the gap between route optimization and motion planning for specific guidance points o...

Safety and performance validation of autonomous agricultural robots is critically dependent on realistic, mobile test environments that provide high-fidelity ground truth. Existing infrastructures focus on either component-level sensor evaluation in fixed setups or system-level black-box testing under constrained conditions, lacking true mobility, multi-object capability and tracking or detecting objects in multiple Degrees Of Freedom (DOFs) in unstructured fields. In this paper, we present a sensor station network designed to overcome these limitations. Our mobile testbed consists of self-powered stations, each equipped with a high-resolution 3D-Light Detection And Ranging (LiDAR) sensor, dual-antenna Global Navigation Satellite System (GNSS) receivers and on-board edge computers. By synchronising over GNSS time and calibrating rigid LiDAR-to-LiDAR transformations, we fuse point clouds from multiple stations into a coherent geometric representation of a real agricultural environment, which we sample at up to 20 Hz. We demonstrate the performance of the system in field experiments with an autonomous robot traversing a 26,000 m² area at up to 20 km/h. Our results show continuous and consistent detections of the robot even at the field boundaries. This work will enable a comprehensive evaluation of geofencing and environmental perception capabilities, paving the way for safety and performance benchmarking of agricultural robot systems.

We propose BiCDO (Bias-Controlled Class Distribution Optimizer), an iterative, data-centric framework that identifies Pareto optimized class distributions for multi-class image classification. BiCDO enables performance prioritization for specific classes, which is useful in safety-critical scenarios (e.g. prioritizing 'Human' over 'Dog'). Unlike uniform distributions, BiCDO determines the optimal number of images per class to enhance reliability and minimize bias and variance in the objective function. BiCDO can be incorporated into existing training pipelines with minimal code changes and supports any labelled multi-class dataset. We have validated BiCDO using EfficientNet, ResNet and ConvNeXt on CIFAR-10 and iNaturalist21 datasets, demonstrating improved, balanced model performance through optimized data distribution.