Indoor Pose Estimation Using Reflective Landmarks and SLAM Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining the state of a localizing apparatus, such as drones in indoor environments, suffer from inaccuracies due to signal attenuation and multi-path effects, especially in complex environments with significant metal presence, and struggle with error accumulation in SLAM systems.
Innovation Solution
A method that involves receiving images of pre-arranged landmarks with known positions and using these images, along with SLAM features, to establish a joint observation model within a state-space model. This model updates the state of the localizing apparatus and the positions of SLAM landmarks, leveraging the known positions of pre-arranged landmarks to reduce error accumulation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If SLAM systems are used for indoor localization, then the system can operate without external signals, but error accumulation occurs as the map expands away from the initial field of view
Solution Approach 1:
The patent implements feedback by using pre-arranged landmarks with known positions as reference points to continuously correct and reset the accumulated errors in SLAM systems. The localizing apparatus periodically detects these landmarks and uses their known positions to adjust and recalibrate the map, preventing error propagation as the system operates over extended periods and areas.
Solution Approach 2:
The patent introduces pre-arranged landmarks as intermediary reference objects between the SLAM system and the external world. These landmarks serve as mediators that provide known position information to anchor the otherwise drift-prone SLAM map, allowing the system to maintain accurate localization without direct external signal input.
2Measurement precision
If outside-in optical localization systems are used, then localization accuracy can be achieved, but the system scales poorly to larger environments requiring many cameras
Solution Approach 1:
The patent enables the localizing apparatus to perform self-localization by equipping it with a camera to detect pre-arranged landmarks in the environment. Instead of requiring external cameras to observe the apparatus, the apparatus independently captures images and processes landmark positions to determine its own state, significantly reducing system complexity while maintaining localization capability.
Solution Approach 2:
The patent inverts the traditional outside-in localization approach by switching to inside-out tracking, where the localizing apparatus carries its own observation device (camera) to detect environmental landmarks. This reversal transforms the system from requiring multiple external cameras to只需要 a single camera on the apparatus, reducing device complexity while maintaining localization functionality.
3Adaptability or versatility
If RF localization technologies are used, then indoor and outdoor localization can be provided, but signal attenuation and multi-path effects limit usability in complex environments
Solution Approach 1:
The patent replaces RF-based localization systems with an optical-based visual localization system using cameras to detect pre-arranged landmarks. This substitution eliminates the signal attenuation and multi-path effects that plague RF systems in metal-rich environments, as optical signals are not affected by electromagnetic interference or signal reflection in the same way, thereby improving reliability in complex indoor environments.
Data Source
Figure 1
Figure 2
AI summary
The invention relates to a method for determining a state xk (9) of a localizing apparatus at a time tk, the state xk being a realization of a state random variable Xk. The method comprises the following steps: a) receiving a first image (1) of a scene of interest (15) in an indoor environment (15), wherein the indoor environment (15) comprises N pre-arranged landmarks (16) having known positions in a world coordinate system (12), N being a natural number; b) receiving a second image (2) of a scene of interest (15) in the indoor environment (15); c) receiving a state estimate Formula Î (3) of the localizing apparatus at the time tk; d) receiving positions of currently mapped simultaneous-localization-and-mapping (SLAM) landmarks (4) in the scene of interest (15), wherein a map state sk comprises at least (i) the state xk of a localizing apparatus, (ii) the positions of the currently mapped SLAM landmarks (4), and (iii) the positions of the pre-arranged landmarks (16); e) determining (5) positions of features in the first image (1), being a natural number smaller than or equal to, and determining (5) an injective mapping estimate from the features into the set of pre-arranged landmarks (16); f) determining (6) positions of L SLAM features in the second image (2), and determining m SLAM features in the L SLAM features, wherein said m SLAM features are related to the n currently mapped SLAM landmarks (4), and determining (6) a SLAM injective mapping estimate from the m SLAM features into the set of the n currently mapped SLAM landmarks (4); g) using the determined injective mapping estimate and the determined SLAM injective mapping estimate to set up (7) a joint observation model as part of a state-space model, wherein the joint observation model is configured to map a map state random variable Sk of which the map state sk is a realization onto a joint observation random variable Zk, wherein at the time tk, an observation zk is a realization of the joint observation random variable Zk, and wherein the observation comprises the position of at least one of the M features in the first image (1) and the position of at least one of the m SLAM features in the second image (2); and h) using (8) (i) the state estimate Formula Î (3), (ii) the joint observation model, and (iii) the observation zk, to determine the state xk (9) of the localizing apparatus at the time tk and to update the positions of the n currently mapped SLAM landmarks. The invention also relates to a computer program product and an assembly.