reading from https://arxiv.org/pdf/2109.13160.pdf
Сравнение производительности алгоритмов локализации и навигации на разных датасетах для разного класса устройств
На рисунке выше показаны значения оценки базовой точности различиных методов. Используются данные простых последовательностей из разных датасетов - нет сложных условий освещения или им подобных.
Как видим на рисунке выше для мобильной платформы Jetson Nano (левая колонка), значения ошибок для определенных алгоритмов и данных выше чем в аналогичных колонках.
Для платформы с меньшей производительностью чем стационарный компьютер или сервер заметно снижение качетсва локализации для некоторых методов: ReF, Ovins.
ReFusion [2] is a dense RGB-D 3D reconstruction method which exploits residuals obtained after the registration of input data with the reconstructed model to identify and filter out dynamic elements in the scene.
OpenVINS [7] is a stereo visual-inertial SLAM system which uses an Extended Kalman Filter to fuse visual odom- etry with inertial measurements.
Baseline performance – We evaluate the trajectory estima- tion accuracy of each SLAM system on selected sequences of widely-adopted datasets where no significant perturbations are present. The RGB-D based SLAM systems are evaluated with 12 sequences from the TUM freiburg1 and freiburg2 datasets [6] and the 4 sequences of the ICL-NUIM living room dataset [5]. Our results (Figures 3-a and 3-b) are consis- tent with the existing literature. ORB-SLAM2, ORB-SLAM3 and ElasticFusion are accurate within 1% on all sequences and no individual runs exceeded 3% error. FullFusion and ReFusion maintained their ATE below 3% on most runs, but scored worse than the aforementioned systems (with few exceptions). ORB-SLAM3 is the most accurate in this baseline setting, with ORB-SLAM2 closely after. SLAM systems supporting stereo and visual-inertial SLAM are evaluated on the 7 easy and medium sequences of the EuRoC-MAV dataset. Figure 3-c shows all 3 SLAM systems have similar accuracies and performed within a 0.5% error margin.
motion, non-uniform illumination, and dynamic scenes. The experiments have covered 6 datasets across 3 computing platforms, in both episodic and long-term operation settings. Thus, this evaluation is the most comprehensive study of the robustness of SLAM systems to date. By including the Nvidia Jetson Xavier platform, we also consider constraints associated with deployments on systems embedded within robots. Overall, we have found that ORB-SLAM3 provides the best balance between baseline accuracy, illumination and fast changes, support for dynamic environments and Lifelong scenarios, although its FPS is below 15 (5 FPS on Jetson). Considering the three dense SLAM systems, FullFusion provides the best balance, but reaches 30 FPS only on the laptop and workstation (Jetson 25 FPS). ElasticFusion offers between 40-50 FPS processing on the three platforms, but its robustness falls below the other SLAM systems. Finally, the sparse SLAM systems have proved more robust than the dense ones, probably because there are fewer data points which can negatively impact pose estimation. We consider that combining sparse tracking with dense 3D reconstruction will help systems build expressive represen- tations while maintaining high robustness.