✳flâneur — a map of the web's best reading

A brief introduction to GraphSLAM | by Shiva Chandrachary | Medium

shivachandrachary.medium.com · 2,284 words · saved by 1 readers

GraphSLAM is a SLAM algorithm that solves the full SLAM problem, i.e, the algorithm recovers the entier path and map instead of just the recent pose and map. This allows it to consider dependencies between current and previous poses. One of the major benefits of the GraphSLAM algorithm is its reduced need for significant onboard processing capability. Unlike FastSLAM, which uses particles to estimate the robot’s most likely pose, GraphSLAM works with all of the data at once to find the optimal solution. Let’s understand how a graph is constructed Any two nodes are connected with an edge, and they are called a soft spatial constraint. It’s called soft because motion and measurement data are uncertain and constraints will have some amount of error present in them. Soft constraints come in two forms: Motion constraints between two successive robot poses and Measurement constraints between a robot pose and a feature in the environment. In the image above, x0 and x1 represent robot poses. T

A brief introduction to GraphSLAM Shiva Chandrachary 10 min read · Jan 12, 2021 -- 1 Listen Share GraphSLAM is a SLAM algorithm that solves the full SLAM problem, i.e, the algorithm recovers the entier path and map instead of just the recent pose and map. This allows it to consider dependencies between current and previous poses. One of the major benefits of the GraphSLAM algorithm is its reduced need for significant onboard processing capability. Unlike FastSLAM, which uses particles to estimate the robot’s most likely pose, GraphSLAM works with all of the data at once to find the optimal sol

Explore this link on the map →

related reading