Abstract
In this paper, we propose a model-free volumetric Next Best View (NBV) algorithm for accurate 3D reconstruction using a Markov Chain Monte Carlo method for high-mix-low-volume objects in manufacturing. The volumetric information gain based Next Best View algorithm can in real-time select the next optimal view that reveals the maximum uncertainty of the scanning environment with respect to a partially reconstructed 3D Occupancy map, without any priori knowledge of the target. Traditional Occupancy grid maps make two independence assumptions for computational tractability but suffer from the overconfident estimation of the occupancy probability for each voxel leading to less precise surface reconstructions. This paper proposes a special case of the Markov Chain Monte Carlo (MCMC) method, the Gibbs sampler, to accurately estimate the posterior occupancy probability of a voxel by randomly sampling from its high-dimensional full posterior occupancy probability given the entire volumetric map with respect to the forward sensor model with a Gaussian distribution. Numerical experiments validate the performance of the MCMC Gibbs sampler algorithm under the ROS-Industry framework to prove the accuracy of the reconstructed Occupancy map and the completeness of the registered point cloud. The proposed MCMC Occupancy mapping could be used to optimise the tuning parameters of the online NBV algorithms via the inverse sensor model to realise industry automation.
Highlights
With the development of industrial robotics and robotic vision, autonomous 3D reconstruction of one-of-a-kind objects for industrial manufacturing attracts more and more interest
The Markov Chain Monte Carlo (MCMC) method can be used to improve the performance of the online two independence assumptions methods by optimising the tuning parameters of the update terms within the inverse sensor model off-line w.r.t to the cross entropy of the benchmark Octree map that is otherwise only exist for toy examples in simulation
The Markov random field image model was analysed to be a lattice system with Gibbs distribution to converge to the maximum a posterior (MAP) estimate and sufficient Gibbs sampler algorithms exist in 2D image processing, the Gibbs sampler has not been extended into the scope of high-dimensional 3D volumetric visual processing, e.g. Octree, 3D reconstruction, RGB-D cameras and etc
Summary
With the development of industrial robotics and robotic vision, autonomous 3D reconstruction of one-of-a-kind objects for industrial manufacturing attracts more and more interest. For the sake of Industry 4.0, as well as the model-free request for manufacturing, this paper proposes a fully autonomous Best View (NBV) solution under the open-source ROS-Industry framework Quigley et al [7], improved by the Markov Chain Monte Carlo (MCMC) method, namely the Gibbs sampler, to generate highly accurate volumetric maps to guide the Best View algorithm to reconstruct surface geometry. – Propose a 3D forward sensor model with Gaussian distribution as the full conditional probability for the Gibbs sampler to generate true probabilistic volumetric maps directly in high dimensional space without any independence decomposition. – Propose a MCMC Gibbs sampler NBV by decomposing the high-dimensional sample space by the full conditional probability of each voxel, to generate benchmark 3D Occupancy maps, for accurate 3D surface reconstruction and optimisation
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
