Sensor Data Damping via Dynamic Threshold Clamping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Sensor data from motion tracking, audio, temperature, and light sensors is noisy due to equipment and measurement noise, making it difficult to accurately control downstream systems like robots and natural user interfaces.
Innovation Solution
A damping logic system that processes sensor data to remove noise by using a clamping mechanism with a dynamically adjustable threshold, based on state data and velocity, to produce low-latency damped data for controlling systems such as game systems and natural user interfaces.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If linear damping is applied to sensor data, then noise and jitter are reduced, but latency increases
Solution Approach 1:
The patent applies dynamic damping where the damping factor is adjusted based on the velocity of sensor data changes. When velocity is high (rapid movements), damping is reduced to maintain responsiveness and reduce latency. When velocity is low (stationary or slow movements), damping is increased to reduce noise and jitter. This dynamic adaptation resolves the contradiction by making the damping strength context-dependent rather than fixed.
Solution Approach 2:
The patent changes the damping parameter based on the velocity of sensor data. By monitoring how quickly sensor values change and adjusting the damping coefficient accordingly, the system optimizes the balance between noise reduction and latency. This parameter adaptation allows the system to maintain precision when needed while preserving responsiveness when motion occurs.
2Measurement precision
If damping is applied to sensor data, then control accuracy is improved, but system responsiveness deteriorates
Solution Approach 1:
The system dynamically adjusts damping strength based on motion velocity. During rapid movements, damping is minimized to preserve system responsiveness and prevent lag. During stationary or slow periods, damping is maximized to improve control accuracy by reducing noise. This dynamic behavior resolves the contradiction between accuracy and responsiveness.
Solution Approach 2:
The damping coefficient is changed as a function of sensor data velocity. This parameter modulation allows the system to switch between high-damping mode (for accuracy during stillness) and low-damping mode (for responsiveness during motion), effectively resolving the trade-off between control accuracy and system responsiveness.
3Measurement precision
If fixed threshold clamping is used, then noise removal is effective, but adaptability to different motion states is lost
Solution Approach 1:
The patent implements a dynamic threshold that adapts to different motion states based on velocity information. Instead of using a fixed threshold, the system adjusts the clamping threshold according to how quickly sensor values are changing. This allows effective noise removal during stationary periods while maintaining sensitivity during rapid movements, thereby resolving the contradiction between noise removal effectiveness and adaptability.
Solution Approach 2:
The clamping threshold parameter is dynamically changed based on the velocity of sensor data. This adaptation allows the system to maintain effective noise removal across varying motion conditions, resolving the contradiction between consistent noise removal performance and adaptability to different motion states.
Data Source
AI summary
Sensor data damping is described, for example, to remove jitter from sensor data to enable control of a computing device. In various examples damped sensor data is compared with a threshold and clamped to the threshold in the case a difference between the sensor data and the damped sensor data is above the threshold. In various examples the damped data has low-latency and is used to control a downstream system such as a game system, natural user interface, robotic system, augmented reality system or other system. In examples, the threshold is adjusted on the fly on the basis of any one or more of: state data from the downstream system, frequency of clamping, velocity of the sensor data. In some examples, the sensor data comprises human/animal joint positions from a depth sensor.