Robot Navigation Bias Estimation for Accurate Route Recreation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Robots face challenges in accurately recreating learned routes due to biases and noise in sensor and odometry data, leading to navigation errors and reduced performance.
Innovation Solution
A system and method for determining biases in robot parameters using probability distribution functions (PDFs) generated from sensor data, allowing for accurate navigation and route recreation by accounting for measurement biases and noise.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If robot uses sensor and odometry data for navigation, then robot can learn and recreate routes, but measurement biases cause navigation errors and reduced accuracy
Solution Approach 1:
The system implements feedback by continuously monitoring sensor and odometry measurements during route execution, comparing actual performance against expected performance, and using this information to identify and correct biases in the measurement instruments. This closed-loop approach allows the robot to adapt to measurement errors and improve navigation accuracy over time.
Solution Approach 2:
The system changes parameters by adjusting the estimated biases of sensor and odometry instruments based on observed navigation errors. By dynamically updating bias parameters and compensation values, the system adapts to drift and calibration errors, transforming the navigation system from static to adaptive parameter adjustment.
2Extent of automation
If robot navigates routes using sensor data, then robot can perform tasks autonomously, but noise in measurements reduces navigation precision
Solution Approach 1:
The system uses feedback to continuously monitor navigation performance and identify patterns in measurement noise. By analyzing accumulated data from multiple route executions, the system distinguishes between random noise and systematic biases, enabling more accurate compensation and maintaining autonomous navigation precision despite noisy measurements.
Solution Approach 2:
The system performs preliminary calibration and bias estimation before autonomous navigation tasks. By pre-characterizing sensor and odometry biases through initial training runs, the system prepares compensation parameters that improve navigation accuracy from the start of autonomous operation, reducing the impact of measurement noise.
3Productivity
If robot executes routes using learned data, then robot can reproduce navigation paths, but biased instruments cause oversteering and incorrect route recreation
Solution Approach 1:
The system dynamically changes instrument bias parameters based on observed execution errors. By adjusting steering angle biases, velocity biases, and position offsets according to accumulated performance data, the system corrects systematic errors that cause oversteering and route deviations, maintaining precision without sacrificing execution speed.
Solution Approach 2:
The system performs self-calibration by automatically identifying and correcting its own measurement biases without external intervention. Through autonomous analysis of navigation errors and self-adjustment of bias parameters, the robot maintains accurate route recreation capability while continuing normal operations, eliminating the need for manual recalibration.
Data Source
AI summary
Systems, apparatuses, and methods for bias determination and value calculation of parameters of a robot are disclosed herein. According to at least one exemplary embodiment, a bias in a navigation parameter may be determined based on a bias in one or more measurement units, wherein a navigation parameter may be a parameter useful to a robot to recreate a route such as, for example, velocity and the bias may be accounted for to more accurately recreate the route and generate accurate maps of an environment.