Testing and validation of autonomous agricultural vehicle trajectory control systems for safety compliance

The service provides specialised support for testing an autonomous vehicle's ability to properly maintain a preset trajectory, enabling it to operate autonomously.

Interested in this service? Contact us at kontakt@agrifoodtef.pl
 

Overview

The service evaluates how a self-moving autonomous robot or vehicle maintains a predefined and AI-controlled trajectory, providing specialised support for this assessment. As a core part of this service, devices and control systems are verified for their adherence to work safety principles. This verification includes analysing the correctness of maintaining the preset trajectory and stopping before the defined limits of the work surface. To achieve this, an advanced testing methodology based on ISO 18497 provisions is utilised, along with specialised equipment to control the vehicle's track maintenance. The support provided is tailored to the customer's specific needs, including preparing the test site according to requirements and providing necessary control equipment and components for proper testing. These activities ultimately lead to increased automation efficiency and precision of controlled objects, particularly in agricultural applications.

More about the service

• Before using the service:Customers own robotic, autonomous devices for fieldwork, technological processes, or other applications within the agriculture and agri-food domains. However, they need support to confirm that these autonomous vehicles maintain current safety standards regarding their driving path and stopping behaviour within designated areas. Without this verification, there's a risk that autonomous operations could lead to collisions with obstacles (such as other objects or trees outside the plot boundary) or even falls off steep plot boundaries.

This uncertainty can hinder the safe and efficient deployment of autonomous vehicles in real-world conditions.• After using the service: Our service assesses the maintenance of safety procedures for the operation of self-driving robots and autonomous vehicles. The service provider offers to prepare and perform safety tests of these vehicles under conditions similar to their expected real-world operation, adhering to a methodology based on ISO 18497 requirements.

◦ The customer will obtain a qualitative assessment indicating whether their vehicle meets current safety standards in terms of maintaining its set driving path and stopping before the boundary of the tested area.
◦ Maintaining these requirements will enable safe autonomous operation (driving) of the vehicle or robot in real conditions, effectively preventing collisions and protecting against incidents such as vehicles falling in areas with steep faults.
◦ If the requirements are not met, the customer will receive information on the probable reasons and suggestions for adjustments to fulfil the assumptions by the tested vehicle.

These activities are expected to increase automation efficiency and precision of controlled objects in agricultural applications.

It is important to note that the service focuses on testing and evaluating compliance with existing standards and assumptions, rather than on developing new solutions or methodologies from scratch, which aligns with the purpose of agrifoodTEF.

Our service is based on a structured course of its implementation, involving the provision of support to ensure testing conditions for the safe movement of autonomous vehicles under conditions in which they perform specific agrotechnical work. The testing conditions will be tailored to the customer's needs in terms of field operations, technological processes, and plants.

The stages of its implementation are as follows:

• Initial consultation: The customer provides a description of the vehicle (test object), the technical and operational parameters of its operation, and its equipment in the form of the machinery or tools it uses in order to determine the scope of the tests and the size of the test areas needed to carry them out. The description should also include the relevant parameters, indicators, and limit ranges to unify the range of input data and their configuration. This stage typically takes from 1 to 5 working days, depending on the complexity of the information provided and the client's requirements.

• Problem analysis: A comprehensive analysis of the problem is carried out, taking into account specific guidelines and expectations. This allows us to jointly determine the detailed methodology for performing the tests and the scope of the support process, relying on existing standards such as ISO 18497. This stage requires active engagement from both parties to establish all details and typically lasts from 3 to 10 working days.

• Planning: A detailed plan is developed defining the scope, methodology, and schedule of activities. This plan is discussed with the customer and approved prior to testing. At this stage, the required resources from both the service provider and the customer are clearly defined, along with the estimated total duration of the entire testing process, which can range from a few days to several weeks depending on the complexity of the tests. The timeframe for this stage depends on the customer’s requirements and is determined on a case-by-case basis.

• Testing and analysis: The testing and analysis processes are carried out in a professional manner using appropriate support equipment. During the physical testing phase, the customer's presence on-site is not mandatory, but it is allowed and even welcome. In practice, the customer or their representative (e.g., technical staff) will typically want to be present to observe the tests and participate in ongoing consultations. The timeframe for this stage aligns with the arrangements defined during the planning phase.

• Data analysis: Collected data is subjected to detailed analysis using advanced methods that allow for comprehensive validation of the correctness of the tested technological process and the device implementing it. The timeframe for this stage depends on the volume of data obtained during the tests, as a larger dataset may require additional time for processing and analysis.

• Reporting: A report is prepared containing synthetic information about the scope, course, and effects of the conducted tests. The report is delivered to the client within an agreed timeframe, usually within 5-10 working days after data analysis completion.

• Post-test support: Post-test support is offered in the form of workshops or consultations for the team, during which the obtained results are discussed.The entire process is flexible and tailored to specific needs, ensuring professionalism and commitment at every stage of cooperation.

Our service is inherently flexible and always tailored to the specific needs of the customer. It is important to emphasise that we focus on testing and evaluating the compliance of autonomous vehicles with existing standards and safety assumptions, rather than on developing new solutions or methodologies from scratch, which aligns with the purpose of agrifoodTEF.

Aspects that are customisable to the customer's needs include:

• Testing methodology: We develop and adapt testing methodologies for autonomous vehicle trajectories. This methodology adheres to the guidelines of the ISO 18497 standard for agricultural robotics and AI, while also incorporating the individual technical and operational parameters of the customer's vehicle, as well as the specifics of the field operations, technological processes, and plants for which it is intended.

• Scope and size of test areas: These are precisely determined based on the vehicle description provided by the customer to reflect real working conditions and ensure appropriate testing environments. The customer's description should include relevant parameters, indicators, and limit ranges to unify input data and configuration.

• Detailed plan and schedule: A detailed plan defining the scope, methodology, and schedule of activities is prepared in close cooperation with the customer, considering their specific guidelines and expectations, allowing for optimal planning of all activities. This plan is discussed with the customer and approved prior to testing.

• Scope of the support process: This is jointly determined with the customer after a detailed problem analysis.Known limitations in service customisation:

• No support for the development of new solutions: The service does not include support for the development of software, hardware, or algorithms from scratch. Instead, it focuses on the validation and assessment of existing control systems to verify their compliance with safety principles and the maintenance of a preset trajectory. The primary purpose of the service is "testing and experimentation".

• Weather conditions: As physical tests are conducted under conditions similar to actual field operations, their execution is dependent on suitable weather conditions. Extreme weather events (e.g., heavy rain, snow, wind, severe frost, or extreme heat) may require rescheduling or modification of planned tests to ensure safety and reliability of results.
Location
Poland
Type of Sector
Arable farming
Horticulture
Livestock farming
Tree Crops
Type of service
Collection of test data
Conformity assessment
Data analysis
Data augmentation
Desk assessment
ELSA assessment
People training
Performance evaluation
Test design
Test execution
Test setup
Accepted type of products
Design / Documentation
Physical system
Software or AI model