Designing a Mixed-Initiative Decision-Support System for Multi-UAS Mission Planning

The United States and its partners have begun a program to extend human presence in the solar system, beginning with a return to the Moon. The Lunar Reconnaissance Orbiter (LRO) is the first step in this endeavor, an unmanned mission to create the comprehensive atlas of the Moon's features and resources necessary to design and build a lunar outpost. LRO focuses on the selection of safe landing sites, identification of lunar resources, and the study of how the lunar radiation environment will affect humans. In order to achieve its objectives LRO carries a payload comprised of six instruments and one technology demonstration. In addition to its scientific mission LRO will use new technologies, systems and flight operations concepts to reduce risk and increase productivity of future missions. As part of the effort to achieve robust and efficient operations the LRO Mission Operations Team (MOT) will use its Mission Planning System (MPS) to manage the operational activities of the mission during the Lunar Orbit Insertion (LOI) and operational phases of the mission. The MPS, based on GMV’s flexplan tool, will receive activity and slew maneuver requests from multiple science operations centers (SOC), as well as from the spacecraft engineers. flexplan will apply scheduling rules to all the requests received and will generate conflict free command schedules in the form of daily stored command loads for the orbiter and a set of daily pass scripts that help automate nominal real-time operations. The MPS will also track the status of onboard flight software load files in the various onboard memory locations. The LRO mission is using a new Flight Software architecture, the Core Flight Executive (cFE), in which all onboard memory is managed as a file system. The MPS will interact with the ITOS real-time system to receive messages whenever changes are made to the memory content. The MPS will further interact with the real-time system in a closed loop mode to monitor the progress of planned activities. The LRO mission will use this feedback to provide continuous visibility of the status of daily operations to all of its team members via flexplan web-based interface, which includes a timeline of upcoming operations. By using a single MPS system to schedule multiple mission elements and generate automation products, the MOT will be able to reduce the risk of operator errors, while maintaining a high level of mission operations awareness among all mission elements. This in turn, achieves safer and more efficient flight operations at low cost.

: Current mission preparation and analysis methods place an undue burden of effort on conventional and special operations forces to effectively synchronize and execute their increasingly complex operational responsibilities in a rapidly changing global environment. This thesis developed a tool for the United States Special Operations Command (USSOCOM) in support of their Mission Planning, Analysis, Rehearsal, and Execution (MPARE) initiative to allow special operations forces commanders and staffs to conduct mission planning and analysis in a distributed environment, and rapidly produce dynamic synchronization matrices and scheduling products. Operations research methods provide the foundation for the analysis. The system developed in this thesis is called the Special Operations Mission Planning and Analysis Support System (SOMPASS). SOMPASS is simple to learn and operate, provides dynamic changes with little effort, and is universal in application. This system has the capability to execute on any hardware platform, operate across any network connection, and expand easily to support additional users and requirements. This thesis provides not only a demonstration of capabilities through a special operations oriented illustrative scenario, but also a working product that can be adapted for use in mission planning and analysis by all units under USSOCOM.

Advanced countries around the world are spurring the development of Unmanned Surface Vehicles (USVs) that can operate autonomously at marine environment. The key enabling technology for such USVs is the mission planning system (MPS) that can autonomously navigate through the harsh waters. The MPS not only has the functions for the navigation, but also has the capabilities, such as obstacle avoidance, malfunction corrections, dealing with unexpected events, return home functions, and many other eventualities that cannot be programmed in advance. The autonomy levels are increasingly moving higher and it is foreseeable that the trend will continue in the future. The main purpose of this paper is the analysis of the MPS onboard the USVs, in terms of the categories, functions, and technological details. Also, we analyze the case study of autonomous mission planning control systems in various fields and introduce the features that constitute the critical functionalities of the mission planning systems.

Increased operational constraints on spacecraft and ground station operations have led to stricter ground software management and better performance throughput. For Mission Planning and Scheduling Systems, the ability to process numerous scheduling and planning input requests from multiple systems, applying high-level logistical rules, and providing export capabilities to all external entities, including the spacecraft itself, must be balanced against system performance. The constraint on the processing time of the schedule comes from a mission-defined need-by time, which is driven by execution or contact time. flexplan, an operational Mission Planning and Scheduling (MPS) system, has addressed the balance of performance and system capabilities. flexplan utilizes a combination of a relational database and soft-algorithm scheduling to maximize system performance. This paper demonstrates the method for the design of the internal database processing and maintenance. Performance methodologies and capabilities are presented to demonstrate software strength in two case studies: The Lunar Reconnaissance Orbiter (LRO) low lunar-orbit science mission currently operational and the Ground Network Schedule System (GNSS) multi-mission, multi-stations study of NASA’s infrastructure.

Mission planning system is the core part of Earth observation satellites ground management, and it is a complex NP hard problem. This paper summarized the advantages and disadvantages of various mathematical models of the planning system today. And papers also analyzed the adaptability and Optimization result of intelligent algorithms in many different situations. Finally, the article gives the current EOS satellite mission planning system's latest developments and trends in the future.

The KOMPSAT-2 (KOrea Multi-Purpose SATellite-2) was launched in July 28, 2006 and has been normally operated. Mission planning system provides the capability to make conflict-free satellite mission schedule by predicting and analyzing satellite orbital events and users’ imaging requests. The amount of satellite mission planning has been increased due to the growing number of image requests. In order to meet the demands of lots of satellite mission planning activities, all mission plans have to be automatically verified without human intervention. Automation concept is also needed because these planning activities have to be done by the pre-defined timeline. To resolve these problems, MPVS(Mission Planning Verification System) has been developed by mission planning team. The MPVS automatically verifies mission plan with satellite contact schedules, resources and constraints. This paper presents the operational concept of the MPVS which verifies mission plan of the KOMPSAT-2. Also design architecture of the MPVS which is composed of the mission planning verification module, command planning verification module, contact schedule generation module, and database including resources and constraints is presented. This paper shows operation results of the MPVS and validity of the KOMPSAT-2 operation.

This paper describes an interactive computer-aiding system for tactical aircraft mission planning. A Multiple-Objective Decision-Making approach has been applied to the tactical mission planning domain. The planner specifies a set of potential flight routes and selects an optimum one as the result of a human-computer dialogue. In this dialogue, the planner iteratively specifies acceptable constraints and desired weights on several mission-related decision objectives. The system provides graphic feedback about the merit of the selected route vis a vis the specified objectives, and allows the planner to tailor the mission route plan to meet explicit decision objectives.

In this paper, a mission planning system is presented that generates mission plans for a group of unmanned aerial vehicles (UAVs) to provide continuous surveillance over an urban area. Given the information of terrain and buildings in the target area, a two-stage approach is employed to solve the problem. In the first stage, a set of camera locations called the vantage set is generated that provides complete coverage of the target area. In the second stage, one or several UAVs are determined to collectively share the vantage set and their individual paths are generated to carry out the continuous surveillance duty. In both stages, evolutionary algorithms (genetic algorithm for vantage set generation and ant colony system for UAV/path planning) are used to search for the optimal solution. During the search, constraints such as the flying capabilities of UAVs and collision avoidance are imposed to guarantee the feasibility of the final result.

A multi-criteria decision support system for multi-UAV mission planning

Trajectory prediction is critical for ensuring the safety, reliability, and scalability of Unmanned Aerial Vehicle (UAV) in urban environments. Despite advances in deep learning, existing methods often struggle with dynamic UAV conditions, such as rapid directional changes and limited forecasting horizons, while lacking comprehensive real-time validation and generalization capabilities. This study addresses these challenges by proposing a gated recurrent unit (GRU)-based deep learning framework optimized through Look_Back and Forward_Length labeling to capture complex temporal patterns. The model demonstrated state-of-the-art performance, surpassing existing unmanned aerial vehicles (UAV) and aircraft trajectory prediction approaches, including FlightBERT++, in terms of both accuracy and robustness. It achieved reliable long-range predictions up to 4 s, and its real-time feasibility was validated due to its efficient resource utilization. The model’s generalization capability was confirmed through evaluations on two independent UAV datasets, where it consistently predicted unseen trajectories with high accuracy. These findings highlight the model’s ability to handle rapid maneuvers, extend prediction horizons, and generalize across platforms. This work establishes a robust trajectory prediction framework with practical applications in collision avoidance, mission planning, and anti-drone systems, paving the way for safer and more scalable UAV operations.

Developing a software system for mission planners at NASA’s Johnson Space Center (JSC) to plan and schedule activities onboard the Space Shuttle or International Space Station (ISS) is a complex task. Due to the scope of the complete user requirements and budget restrictions, a full development of all capabilities was not practical. The planning system had to include not only capabilities for planning the daily activities onboard the spacecraft, but also had to include the capability to upload the plans to the onboard system and to merge plans and activities from planners at other remote planning centers, including Marshall Space Flight Center (MSFC) and International Partners. The Integrated Planning System (IPS) Consolidated Planning System (CPS) was developed in an incremental fashion to continually meet the most immediate needs for the users. CPS has been under development for over 8 years and has been operational at JSC for 5 years, supporting Space Shuttle Program (SSP) and ISS mission planning. The journey to this point of software capabilities has been slow, but steady. The current size of the software application is over 380 K SLOC. Initial development focused on capability for scheduling SSP missions, as this was the most immediate need. Next, the development focused on ISS capability, such as plan upload and advanced long-range planning techniques. Finally, the development focused on including/merging data from MSFC and the International Partners. This paper documents some of the “Lessons Learned” during the development of the CPS. Some examples of these include: • Focusing on current operational needs creates system requirements that may not be applicable in future operations • File exchange mechanism uses a proprietary file format that is not flexible to change • Size and quantity of current planning data seriously hampers performance of the system • Incremental approach to development adds new functionality as peripheral to core functionality The nature of “Lessons Learned” tends to focus our attention on deficiencies in the software. We look at actions or functionality that could have been implemented better or planned better. Focusing on these may give the reader the false impression that the CPS is a short sighted, unusable system. However, the system is stable and quite comprehensive in its capabilities for planning current ISS and SSP operations. The incremental approach allowed the development team to tailor the software to shifting user requirements as planning user operations scenarios and requirements changed. Overall, the CPS is an excellent planning system. This paper is intended to provide some assistance in a “path through the mine field” of software development for planning space mission operations. It is hoped that these “Lessons Learned” can be applied in future space mission planning system projects to avoid some of the limitations of the current system.

The European Space Operations Centre (ESOC) is responsible for operating spacecraft covering a large range of mission types. These range from planetary probes/deep space missions through observatories and science missions to earth observation. All of these different types of missions have the need to plan operations on the spacecraft and, increasingly, of the ground segment. Typically the planning systems used to carry out this process have been implemented on a mission specific basis, at best utilizing some elements from a similar mission type. Reuse across different mission types has typically been almost non-existent. In order to increase synergy in the development of mission planning systems ESOC has started work to define a mission planning framework that can be used to build planning systems for any of the mission types operated. To do this it has been necessary to obtain input from the flight control teams of missions in operation and development. This has proved to be an interesting exercise as one of the main issues has been defining the terminology that was required. Whilst the terminology used amongst missions of the same type is similar it varies significantly between the mission types. This obviously adds considerably to the difficulty in trying to define a common approach to mission planning. To address this issue it was first necessary to define a set of terminology that was clear to all participants and could be used as the basis for defining a common concept. Once this has been done it then proved possible to identify a workflow that was largely applicable to most (although not all) missions. This workflow is as seen by the “planning team” part of a flight control team (FCT); no assumptions have been made as to whether the planning team is solely dedicated to planning activities, only that the planning is carried out as an FCT function.

Considering emerging technologies of the intelligent space and wireless sensor network, a business architecture, system architecture and technology architecture are proposed for the mission planning system of a space sensor network. The business architecture of a mission planning system is constructed from application flows, such as those of task modelling, planning and scheduling. The system architecture is designed from the software platform, resource configuration and node structure. The technology architecture is proposed according to applied key technologies. Key technologies are also proposed for the mission planning system of a space sensor network. Through the strengthening of top-level planning and the innovation of the conception design, the exploration of the architecture and key technologies of the mission planning system of a space sensor network will promote technological innovation and provide comprehensive benefits.

The European Space Operations Centre (ESOC) is currently carrying out an ambitious program to standardize the infrastructure used throughout its ground segments in an effort to improve the reliability, cost effectiveness and interoperability. The end product of this program will be the ESA Ground Operation System (EGOS). This paper summarizes the program rational, its goals, the architecture, development approach and current status. I. Introduction A. Overview SOC is responsible for the provision of ground segments for the operation and control of many ESA missions. The scope of the ground segment encompasses the elements required on ground to provide the requisite monitoring and control capabilities. This ranges from the groundstations, through the mission control and flight dynamics functionality, to the archiving and distribution of data to the science community and, in some cases, the initial processing of science data. The provision of simulators for training and operations validation purposes is also covered. Many of the systems that are used in the ground segment have been developed in the past using a variety of different technologies and policies, quite often as standalone projects which did not take into account other developments. EGOS is therefore intended to provide a more standardised infrastructure, which encourages systematic reuse and thus reduce development and maintenance costs. The approach adopted is to use well established S/W engineering principles, build on existing systems and experience, use international standards where available and avoid proprietary systems/standards where possible. Initially EGOS is concentrating on the areas of mission control systems, simulators and some of the ground station equipment, in particular the base-band systems and ground station monitoring and control. In the longer term it is envisaged that the scope will be extended to cover satellite electrical ground support equipment (EGSE), flight dynamics and mission planning systems. EGOS must also permit easy expandability, as it is not possible to forecast with certainty future system requirements. The approach adopted in the architecture is one of a service oriented system based on component application development. The service and component architectures represent two views of the system and are therefore not mutually exclusive, but complementary. Within the overall architecture categories of components have been identified. The first of these are the Core components, these provide the essential services and functionality on which the other components are built. Next come the Common components, these provide widely used services such as file archiving, user management etc. The Core and Common components can effectively be thought of as middleware for the ground data systems. The third category, Custom components, consists of the applications that implement the end user needs. These map roughly onto subsystems, e.g. telemetry chain, telecommanding chain, onboard software maintenance system etc. Encapsulating the overall EGOS architecture is the Service Management Framework (SMF). This provides the mechanism through which the underlying systems expose services to external users and systems in a well defined and controlled manner. To avoid unnecessary dependencies between a set of rules has been prepared specifying whether the different categories of components may utilise the services provided by other components categories. * Head of the Ground Station Backend Section, OPS-GIB † Senior Software Engineer, Data System Infrastructure Division, OPS-GI ‡ Head of the Product Maintenance & Support Section, OPS-GIM § Head of the Operation Centre System Infrastructure Section, OPS-GIC ** Head of the Data System Infrastructure Division, OPS-GI

In order to improve the capability of aircraft flight test data, this paper proposes an intelligent flight test mission planning system which is based on a big data platform, combined with artificial intelligence algorithms, modeling and simulation technologies. This system can meet the flight test requirements of the new aircraft, and complete the test flight syllabus, decompose and integrate the test flight task, make the test flight plan for the specific sortie and verify the simulation. Finally, it can formulate a scientific and efficient test flight plan. The intelligence of the system runs through the entire workflow, including the design, validation and finalization of the flight test plan. This article will introduce the main principles and methods of this system.