Multi-Domain Mission Path Planning With Physics-Based Genetic Search
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing path planning systems for multi-domain assets require significant human intervention, are not goal-based, and fail to accurately consider physics constraints, making them inefficient and imprecise, especially in multi-domain scenarios.
Innovation Solution
A multi-domain, goal-based path planner utilizing a genetic algorithm backed by physics models, which automatically generates mission plans for multiple assets across different domains without a user interface, allowing for parallel processing and accurate trajectory planning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional human intervention methods are used for path planning, then trajectory precision can be maintained, but the time required increases significantly
Solution Approach 1:
The patent replaces the mechanical human operator system with an automated path planning system that uses algorithms to generate and evaluate paths. The system automatically computes trajectories considering vehicle physics models, constraints, and goals without requiring human verification of each path detail, thus maintaining precision while dramatically reducing planning time.
Solution Approach 2:
The path planning system performs self-verification by automatically checking generated paths against vehicle constraints, physics models, and mission goals. The system validates its own outputs through automated scoring and filtering mechanisms, eliminating the need for external human verification while maintaining high trajectory precision.
2Duration of action of moving object
If waypoint-based control is used, then longer future routes can be covered, but navigation accuracy to waypoints deteriorates
Solution Approach 1:
The system transforms discrete waypoint targets into continuous trajectory definitions with temporal parameters. Instead of specifying only position coordinates, the system defines complete path trajectories with speed, acceleration, and timing information, allowing vehicles to follow smooth paths with high accuracy while covering extended routes.
3Measurement precision
If vector control with real-time operator input is used, then dynamic trajectory control precision is improved, but operator mental burden increases
Solution Approach 1:
The system replaces the human operator's continuous control inputs with an automated control system that generates and executes trajectories autonomously. The automated system maintains precision by using vehicle physics models and constraint satisfaction algorithms, while eliminating the need for sustained operator attention and reducing mental workload.
4Measurement precision
If manual path creation is used, then path accuracy can be ensured, but the system cannot support multiple domain requests in rapid succession
Solution Approach 1:
The system replaces manual path creation with automated algorithmic generation that can process multiple requests in parallel. The automated system maintains path accuracy through physics-based validation and constraint checking while achieving high throughput by eliminating sequential manual operations and enabling concurrent path planning for multiple vehicles and domains.
Solution Approach 2:
The system pre-computes and caches vehicle physics models, constraint parameters, and domain-specific information before requests arrive. This preliminary preparation enables rapid processing of multiple requests by reusing pre-loaded data and models, maintaining accuracy while significantly increasing processing throughput.
5Device complexity
If prior path planning systems are used, then simple single-domain paths can be generated, but multi-domain scenarios with physics constraints cannot be adequately handled
Solution Approach 1:
The system implements a universal path planning framework that handles multiple domains (aerial, maritime, ground, space) through a common architecture. The system uses domain-agnostic vehicle physics models and constraint representations that can be configured for different vehicle types and operating environments, enabling multi-domain applicability while maintaining manageable system complexity through modular design.
Data Source
AI summary
A computer architecture includes an application program interface (API). The API does not include a user interface. The computer architecture asynchronously receives into the API data relating to mission plan domains from clients. The data include an identification of vehicles, goals of the vehicles, and threats to the vehicles. The mission plan domains include an air domain, a sea or ocean domain, and a land domain. The computer architecture uses a parallel processing scheme to process the mission plan domains from the clients for determining goal priorities for each of the plurality of vehicles, processing the data using a genetic algorithm and physics models associated with the plurality of vehicles, and transmitting to the vehicles path commands based on the processing of the genetic algorithm.