The concept of continuous-time trajectory representation has brought increased accuracy and efficiency to multi-modal SLAM.
However, regardless of these advantages, its offline property caused by the requirement of a global batch optimization is critically hindering its relevance for real-time and life-long applications.
We recently introduce a dense map-centric SLAM method based on a continuous-time representation to cope with this problem.
The proposed system locally functions in a similar fashion to conventional Continuous-Time SLAM (CT-SLAM). However, it removes global trajectory optimization by introducing map deformation.
The computational complexity of the proposed approach for loop closure does not depend on the operation time, but only on the size of the space it explored before the loop closure (Space bounded instead of Time bounded).
It is therefore more suitable for long term operation compared to conventional CT-SLAM. Furthermore, the proposed method reduces uncertainty in the reconstructed dense map by using probabilistic surface element (surfel) fusion.
We demonstrate that the proposed method produces globally consistent maps without a global batch trajectory optimization, and effectively reduces LiDAR noise by fusion in dense surfel mapping.
For more information please contact: Dr Peyman.Moghadam [at] csiro.au
Park, C., Moghadam, P., Kim, S., Elfes, A., Fookes, C., & Sridharan, S. (2017). Robust Photogeometric Localization over Time for Map-Centric Loop Closure. To appear in proceedings of the IEEE Robotics and Automation Letters, Jan 2019.
Park, C., Moghadam, P., Kim, S., Elfes, A., Fookes, C., & Sridharan, S. (2017). Elastic LiDAR Fusion: Dense Map-Centric Continuous-Time SLAM. To appear in proceedings of the IEEE International Conference on Robotics and Automation (ICRA), 1206-1213, 2018.
Park, C., Kim, S., Moghadam, P., Fookes, C., & Sridharan, S., Probabilistic Surfel Fusion for Dense LiDAR Mapping, The IEEE International Conference on Computer Vision (ICCV), 2017, pp. 2418-2426.