Use Cases from our Project Partners

Warehouse Logistics

Magazino is an innovative company producing robots for warehouse intra-logistics.

In Magazino, each robot is given a prioritized list of jobs to perform, and each of them is associated with a hand-written plan, encoded using a behavior tree formalism.

This solution has a few drawbacks: handwriting of the behavior trees requires expert knowledge and quite some brain power to be carried out without mistakes. This makes the writing tedious and the maintenance and update complex and error prone. For example, in addition to take into account the possible failures and unexpected events, such plans need to be crafted considering that the executor might get restarted at any point in time, losing and thus needing to reconstruct the world and internal states. Finally, the behavior trees are typically rather complex and they are not suitable to be analyzed or manipulated by some high-level reasoning mechanism.

In this context, planning techniques can be used to guide the user to design the complex plans required to accomplish the tasks the robot is given, and to follow their execution taking into account the automatic recovery from unforeseen circumstances, e.g., after a failure has occurred.

Production Planning

Meritor HVS AB in Lindesberg manufactures rear and front axles for heavy vehicle applications. The plants comprise both machining and assembly. The order horizon for assembly as well as the assembly process itself are short which allows the planning process to meet the customer needs.

For the machining department we have longer lead-times which forces us to consider order forecasts for planning. Together with a complex machine park with numerous different process steps and different machines with different cycle times depending on the part produced, this makes planning hard.

Today the planner operates day by day, using data on the availability of parts and comes up with a production plan based on what we know at this time. However this is a very time-consuming process and in reality the result are almost every time wrong as changes such as machine breakdowns or unaccomplished target cycle times have occurred during the planning process. This leads to sub-planning activities in the machining departments that focus on the individual department only and as a consequence to an imperfect overall exploitation of resources.

We are expecting as an major end result a tool that provides decision options for the central planners, enabling us to control the production in real-time based on accurate data.

Planning for Space

The ‘Planning for Space’ use case targets the automation of the planning process in the context of multi-asset human-robotic missions as prepared by the National Space Agencies, the European Commission and the European Space Agency. Typical examples are the ExoMars mission for Mars exploration, the Mars Sample Return mission as well as the HERACLES mission for moon exploration and exploitation.

In this context, the ground operators prepare Activity Plans to be executed by the robotic system on the planet as a logical and temporal composition of robotic Activities. In addition, the system shall merge Activity Plans proposed by different remote operators to a final consolidated Activity Plan to be uploaded for execution. The automatisation of this process using the AIPlan4EU framework is expected to decrease the duration of the on-ground planning phase and create more efficient plans in terms of energy consumptions and science return.

Underwater operations

HyDrone is a technology program to develop new concept underwater vehicles, highly reconfigurable with different payloads, expected to carry out underwater missions at different degrees of autonomy and to operate resident underwater without maintenance for up to two years. Typical missions range from geographically restricted inspections and operations on wells, pipelines and processing systems (using wired communication and power), to medium range missions (based on underwater wifi communication and battery power), to long-range missions.

AI planning techniques may help to overcome several problems like:

  • Management of system resources even in abnormal or unexpected conditions happening during mission execution

  • Management of critical/abnormal situations that may lead to damage or risks in general

  • Making the mission’s definition independent from specific field, so allowing to develop a much more general code in the lowest levels of the architecture (improving reusability, testability and development of additional features for the system and not for a specific scenario)

  • Allow planning missions that foresee use of manipulator and tools in general in a seamless way with the rest of the mission

  • Exploiting opportunities that may not be exploited by standard, rigid plans programmed by humans

  • Help in validating missions proposed/developed by humans

Campaign-Planning for Silage Maize Harvesting

Agriculture tech has become a highly digitized, high-technology branch involving powerful and expensive machinery. The need for effective and efficient processes has increased due to economic and ecological constraints. In that situation, automation, optimization and planning methods are natural choices, and they can build on existing high levels of digitization at modern farms and contractors.

One problem class is campaign planning for agricultural processes. It consists of various coordination and optimization problems on several levels of detail. Campaign planning is needed in harvesting processes, which involve harvesters, overloading and transport vehicles, and possibly other resources and agents, depending on the type of crop; it is also of help in spraying chemicals or manure.

The harvesting processes for the various crops, e.g., silage maize, wheat, forage, or sugar beet, vary due to different general requirements and machinery. A particularly complex process is silage maize harvesting, which is the focus of this use-case. The chopping, transport, and compaction of the crop in the silo must be coordinated so that the machines are used to full capacity and downtimes are avoided. Efficiency is important here, as there are requirements in terms of time, economy, and sustainability: When the crop is ripe there is only a limited time window to harvest it. Here, weather conditions can lead to further restrictions or deadlines. Efficiency is also important from an economic point of view, as the machines and labour are expensive and limited. Finally, sustainability aspects such as a reduction in fuel consumption or soil compaction risks could also play a role.

In this use case AIPlan4EU aims to use AI planning techniques in order to plan and optimize the overall silage maize harvest campaigns. This also involves monitoring and adapting plans during the harvest in order to react to changed conditions or deviations from the original plan.