Fleet Navigation State Estimation Using Moving Vehicle References
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing navigation systems struggle to efficiently estimate the state of a fleet of vehicles, particularly when carriers surrounding the main vehicle are moving, as they require specific binary operations that are not easily genericizable, leading to divergence issues with extended Kalman filters.
Innovation Solution
A method utilizing an invariant Kalman filter with a binary operation that combines term-by-term composition of rigid transformations for estimating navigation states in a fleet of vehicles, where the main vehicle uses relative kinematic data from surrounding vehicles as movable reference points, stabilizing the estimation error and simplifying calculations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If an extended Kalman filter is used to estimate the navigation state of a fleet of vehicles, then the system can handle non-linear dynamics, but the estimation error diverges due to lack of error autonomy
Solution Approach 1:
The navigation state is segmented into two independent parts: the state of the main vehicle and the relative states of surrounding vehicles. This segmentation allows the error dynamics to be decoupled, making the estimation error autonomous and preventing divergence while maintaining non-linear handling capability.
Solution Approach 2:
The patent introduces an intermediary coordinate system (the main vehicle's reference frame) that mediates between the global reference frame and the individual vehicle frames. This intermediary frame enables the formulation of relative states that lead to autonomous error dynamics, resolving the contradiction between non-linear adaptability and error stability.
2Reliability
If specific binary operations are designed for fleet navigation estimation, then the invariant Kalman filter can be implemented, but the complexity increases due to non-genericizable operations
Solution Approach 1:
The patent defines universal binary operations for composing rigid transformations that work generically for any fleet configuration. These operations combine translation and rotation in a standardized way, making the invariant Kalman filter implementable without designing specific operations for each scenario, thus reducing complexity while maintaining reliability.
3Adaptability or versatility
If surrounding vehicles are used as movable reference points, then navigation assistance can be provided in dynamic environments, but the estimation becomes more complex due to relative motion
Solution Approach 1:
The patent merges the estimation of the main vehicle's state with the relative states of surrounding vehicles into a unified navigation state vector. This combining approach allows the system to use movable reference points effectively while managing complexity through a structured state representation that the invariant Kalman filter can process efficiently.
Data Source
AI summary
A method for assisting the navigation of a fleet of vehicles including main vehicle and a secondary vehicle movable relative to the main vehicle includes receiving data acquired by one or more sensors, the received data including relative kinematic data between the main vehicle and the secondary vehicle, and estimating a navigation state of the fleet of vehicles by an invariant Kalman filter using the received data as observations. The navigation state includes first variables representative of a first rigid transformation linking a frame attached to the main vehicle to a reference frame, and second variables representative of a second rigid transformation linking the frame attached to the main vehicle to a frame attached to the secondary vehicle. The invariant Kalman filter uses as binary operation an operation including a term-by-term composition of the first rigid transformation and of the second rigid transformation.