Robust pose graph optimization using stochastic gradient descent

Robust SLAM methods can allow robots to recover correct maps even in the presence of incorrect loop closures. While these approaches improve robustness to outliers, they are susceptible to getting caught in local minima, a problem which is exacerbated by poor initial estimates. In this paper, we describe a stochastic gradient descent optimization approach that exhibits greater robustness to poor initial estimates. Our approach can either be used as a stand-alone optimization system or in conjunction with existing methods such as Gauss-Newton solvers. Using a combination of synthetic and real-world datasets, we demonstrate that our proposed approach is able to recover correct pose graphs significantly more frequently than other methods when large initialization errors are present.

Authors

• John Wang
• Edwin Olson

Keywords

• Robustness
• Simultaneous localization and mapping
• Optimization
• Noise
• Convergence
• Stochastic processes
• Gradient Descent
• Stochastic Gradient Descent
• Pose Graph
• Pose Graph Optimization
• Local Minima
• Poor Estimation
• Loop Closure
• Null Hypothesis
• Optimization Problem
• Learning Rate
• State Space
• Mixture Model
• Real-world Data
• Rigid Body
• Lowest Error
• Fisher Information
• Correct Amount
• Gaussian Components
• True Solution
• Learning Rate Schedule
• Gauss-Newton Method
• Stochastic Gradient Descent Method