Lei Sun

Reconstructing geometry, appearance, and motion from videos is a central problem in computer vision and world modeling. In the monocular dynamic setting, this problem is severely ill-posed because the underlying 3D structure and time-varying motion must be inferred from sparse and self-motion views. Mitigating this ill-posedness requires a representation that is both temporally coherent and sufficiently constrained for monocular optimization. We introduce Gaussian Sequences, a framework that converts a monocular video into a temporally coherent sequence of 3D Gaussian primitives for dynamic reconstruction and 3D tracking. At the core of the framework is a global-local SE(3) dynamic field. Global motion patterns provide a compact representation of coherent scene-level dynamics, while local residual motion controllers add bounded corrections for fine non-rigid deformation. To further reduce monocular ambiguity, we leverage foundation-model priors as initialization cues and complementary constraints that guide geometry and motion recovery under weak observations. As an application, Gaussian sequences serve as spatio-temporal observations for differentiable particle simulation, enabling physical parameter identification from monocular videos and action-conditioned dynamics prediction. Experiments on real-world monocular videos show that our Gaussian sequences outperform strong dynamic reconstruction baselines in novel-view synthesis and point tracking. Downstream particle simulation of deformable objects further validates that the reconstructed sequences provide useful information for predictive physical modeling.

Most of the existing LiDAR-inertial odometry (LIO) methods are based on frame-to-map constraints, which may lead to inconsistency in state estimation. Approaches relying on frame-to-frame could address the inconsistency, but easily fail during association loss. In this article, HCF-LIO, a LIO algorithm based on a voxel-map-based hybrid constraint framework is proposed. The approach synergistically integrates single-frame absolute pose constraints (frame-to-map) with multiframe relative constraints (frame-to-frame) within a filtering architecture, effectively balancing short-term accuracy and long-term consistency. In addition, an implicit multiframe constraint approach based on probabilistic voxel map enables plane feature association without explicit tracking. Furthermore, a secondary motion compensation module is implemented after state update, enhancing the accuracy of the LiDAR point cloud map and the quality of multiframe association. To evaluate the proposed framework, extensive experiments and ablation studies are conducted on public datasets and private self-collected dataset. The experimental results demonstrate that HCF-LIO outperforms the state-of-the-art approaches in terms of accuracy and consistency while maintaining real-time operation.

This paper proposes a novel multisensor fusion ground SLAM system that integrates LiDAR, visual, IMU, and wheel encoder data to enhance localization and mapping accuracy in complex environments. The proposed system consists of a LiDAR subsystem and a visual subsystem. To fully leverage the complementary advantages of LiDAR and visual sensors, the two subsystems provide each other with initial values for optimization. LiDAR supplies high-precision depth information to the visual subsystem, while the visual subsystem assists the LiDAR module in motion distortion correction, thereby enhancing the overall accuracy of state estimation. In addition, a novel plane extraction method is proposed based on adaptive dynamic distance thresholds and normal vector consistency verification. And the planar factor is introduced into the LiDAR subsystem for joint optimization. Constraining continuous planar structures (such as the ground and walls) reduces matching ambiguity, which further decreases local map noise and improves mapping accuracy. The proposed method has been extensively evaluated in complex urban campus scenarios, and the results show the effectiveness and accuracy of the algorithm.

A “planning and control” scheme with hysteresis compensation is developed in this article for a compliant grinding device (CGD) driven by a pneumatic actuator with asymmetric full-state constraints. First, in the planning part, we propose an improved hysteresis model and construct its inverse model through the modified inverse multiplicative structure method to deal with the hysteresis nonlinearity between air pressure and contact force in CGD. In the control part, a neural state observer (SO) is developed to observe the rate of change of air pressure by combining a neural network (NN) and an SO. By fusing an NN and a disturbance observer (DO), a neural DO is presented to observe external disturbances. Asymmetric full-state constraints are realized without the feasibility conditions (FCs) of the virtual controller by the fusion of the backstepping design process and nonlinear state-dependent transformations. It is proven that the air pressure closely tracks the planned desired air pressure, as the tracking error of the air pressure is uniformly ultimately bounded. Finally, hardware experiments show that the force control accuracy of the proposed method is within 2N, which achieves a satisfactory force control effect and verifies the effectiveness and robustness of the presented method.

Constructing a globally consistent high-precision map is essential for the application of mobile robots. Existing optimization-based mapping methods typically constrain robot states in pose space during the graph optimization process, without directly optimizing the structure of the scene, thereby causing the map to be inconsistent. To address the above issues, this paper presents a three-dimensional (3D) LiDAR mapping framework (i.e., BA-CLM) based on LiDAR bundle adjustment (LBA) cost factors. We propose a multivariate LBA cost factor, which is built from a multi-resolution voxel map, to uniformly constrain the robot poses within a submap. The framework proposed in this paper applies the LBA cost factors for both local and global map optimization. Experimental results on several public 3D LiDAR datasets and a self-collected 32-line LiDAR dataset demonstrate that the proposed method achieves accurate trajectory estimation and consistent mapping.

This paper proposes a “planning and control” scheme for a compliant mechanism (CM) based on a pneumatic actuator (PAC) with hysteresis nonlinearity and full-state constraints. In the planning part, a novel direct inverse model is presented to compensate for hysteresis nonlinearity, enabling more accurate planning of the desired air pressure based on the desired contact force. In the control part, by fusing fuzzy logic systems (FLSs) and an extended state observer (ESO), a fuzzy ESO is developed to observe the external disturbance and the rate of change of the air pressure. Additionally, the challenges in the controller design caused by full-state constraints are overcome by constructing barrier Lyapunov functions (BLFs). It is proved that all signals of the closed-loop system are bounded, and the tracking error of the air pressure can converge to a small neighborhood of the origin. Finally, the effectiveness and robustness of the proposed method are verified by hardware experiments, which also show that the root mean square errors of force control accuracies are within 2N, achieving satisfactory force control effects.

Constructing a globally consistent high-precision map is essential for the application of mobile robots. Existing optimization-based mapping methods typically constrain robot states in pose space during the graph optimization process, without directly optimizing the structure of the scene, thereby causing the map to be inconsistent. To address the above issues, this paper presents a three-dimensional (3D) LiDAR mapping framework (i.e., BA-CLM) based on LiDAR bundle adjustment (LBA) cost factors. We propose a multivariate LBA cost factor, which is built from a multi-resolution voxel map, to uniformly constrain the robot poses within a submap. The framework proposed in this paper applies the LBA cost factors for both local and global map optimization. Experimental results on several public 3D LiDAR datasets and a self-collected 32-line LiDAR dataset demonstrated that the proposed method achieves accurate trajectory estimation and consistent mapping.

Traditional industrial robots often face challenges in achieving a perfectly polished surface on a workpiece because of their high mechanical rigidity. The active compliance force control device installed at the robotic arm’s end enables high-precision contact force control between the grinding tool and the workpiece. However, the complex hysteresis nonlinearity between cylinder air pressure and output force, as well as various random disturbances during the grinding process, can affect the accuracy of the contact force and potentially impact the grinding effect of the workpiece, even causing irreversible damage to the surface of the workpiece. Given the complex random variation of cylinder output force in the actual grinding process, a rate-dependent hysteresis model based on diagonal recurrent neural network and Pradtl–Ishlinskii models named dRNN-PI is designed to compensate for the complex nonlinear hysteresis of the cylinder and calculate the desired air pressure to maintain a steady contact force on the workpiece. The proxy-based sliding mode control (PSMC) is utilized to quickly track the desired air pressure without overshooting. This paper also proves the controller’s stability using the Lyapunov-based methods. Finally, the accuracy of the proposed hysteresis compensation model and the effectiveness and robustness of the PSMC are verified by experiment results.