In open-pit mining, autonomous trucks are essential for enhancing both safety and productivity. Object detection technology is critical to their smooth and secure operation, but training these models requires large amounts of high-quality annotated data representing various conditions. It is expensive and time-consuming to collect these data during open-pit mining due to the harsh environmental conditions. Simulation engines have emerged as an effective alternative, generating diverse labeled data to augment real-world datasets. However, discrepancies between simulated and real-world environments, often referred to as the Sim2Real domain shift, reduce model performance. This study addresses these challenges by presenting a novel semi-supervised domain adaptation for object detection (SSDA-OD) framework named Adamix, which is designed to reduce domain shift, enhance object detection, and minimize labeling costs. Adamix builds on a mean teacher architecture and introduces two key modules: progressive intermediate domain construction (PIDC) and warm-start adaptive pseudo-label (WSAPL). PIDC builds intermediate domains using a mixup strategy to reduce source domain bias and prevent overfitting, while WSAPL provides adaptive thresholds for pseudo-labeling, mitigating false and missed detections during training. When evaluated in a Sim2Real scenario, Adamix shows superior domain adaptation performance, achieving a higher mean average precision (mAP) compared with state-of-the-art methods, with 50% less labeled data required, achieved through active learning. The results demonstrate that Adamix significantly reduces dependence on costly real-world data collection, offering a more efficient solution for object detection in challenging open-pit mining environments.

Unsupervised Domain Adaptation for Object Detection (UDA-OD) aims to adapt a model trained on a labeled source domain to an unlabeled target domain, addressing challenges posed by domain shifts. However, existing methods often face significant challenges, particularly in detecting small objects and over-relying on classification confidence for pseudo-label selection, which often leads to inaccurate bounding box localization. To address these issues, we propose a novel UDA-OD framework that leverages scale consistency (SC) and Temporal Ensemble Pseudo-Label Selection (TEPLS) to enhance cross-domain robustness and detection performance. Specifically, we introduce Cross-Scale Prediction Consistency (CSPC) to enforce consistent detection across multiple resolutions, improving detection robustness for objects of varying scales. Additionally, we integrate Intra-Class Feature Consistency (ICFC), which employs contrastive learning to align feature representations within each class, further enhancing adaptation. To ensure high-quality pseudo-labels, TEPLS combines temporal localization stability with classification confidence, mitigating the impact of noisy predictions and improving both classification and localization accuracy. Extensive experiments on challenging benchmarks, including Cityscapes to Foggy Cityscapes, Sim10k to Cityscapes, and Virtual Mine to Actual Mine, demonstrate that our method achieves state-of-the-art performance, with notable improvements in small object detection and overall cross-domain robustness. These results highlight the effectiveness of our framework in addressing key limitations of existing UDA-OD approaches.

Visual-inertial odometry (VIO) has been widely applied in mobile robots due to its low cost and versatility. However, the planar motion of mobile robots leads to the degradation of observability in VIO, while slope variations in outdoor environments invalidate the global planar assumption of existing VIO methods, resulting in a reduction in localization accuracy. To address these issues, this paper proposes a novel depth-assisted tightly coupled optimization framework for visual-inertial-wheel odometry. By extracting the local ground as a key reference plane and incorporating the geometric relationship between the mobile robot and the local ground, novel height and attitude constraints are proposed, enabling the framework to effectively suppress drift in the vertical direction as well as in the roll and pitch degrees of freedom. First, voxel filtering is applied to downsample the dense 3D point cloud generated by the depth camera, which reduces computational complexity while removing noise points from the raw point cloud. Subsequently, the RANSAC plane fitting method is employed to robustly identify local ground planes in environments with noise and dynamic objects. Finally, height and attitude constraints are designed based on the local ground plane parameters, and these are integrated with visual reprojection constraints and inertial-wheel preintegration constraints into the tightly coupled optimization framework to further improve pose estimation accuracy. We evaluate the performance of the proposed method using the KAIST Complex Urban Dataset and real-world experiments, and compare it with state-of-the-art visual-inertial methods such as VINS-Fusion and VIW-Fusion. The results demonstrate that the proposed method achieves high accuracy in mobile robot localization and exhibits greater robustness in uneven terrains and dynamic urban environments.