3D scene understanding plays a vital role in vision-based autonomous driving. While most existing methods focus on 3D object detection, they have difficulty describing real-world objects of arbitrary shapes and infini...
3D scene understanding plays a vital role in vision-based autonomous driving. While most existing methods focus on 3D object detection, they have difficulty describing real-world objects of arbitrary shapes and infinite classes. Towards a more comprehensive perception of a 3D scene, in this paper, we propose a SurroundOcc method to predict the 3D occupancy with multi-camera images. We first extract multi-scale features for each image and adopt spatial 2D-3D attention to lift them to the 3D volume space. Then we apply 3D convolutions to progressively upsample the volume features and impose supervision on multiple levels. To obtain dense occupancy prediction, we design a pipeline to generate dense occupancy ground truth without expansive occupancy annotations. Specifically, we fuse multi-frame LiDAR scans of dynamic objects and static scenes separately. Then we adopt Poisson Reconstruction to fill the holes and voxelize the mesh to get dense occupancy labels. Extensive experiments on nuScenes and SemanticKITTI datasets demonstrate the superiority of our method. Code and dataset are available at https://***/weiyithu/SurroundOcc.
3D scene understanding plays a vital role in vision-based autonomous driving. While most existing methods focus on 3D object detection, they have difficulty describing real-world objects of arbitrary shapes and infini...
详细信息
The limitations of conventional animal tests and two-dimensional cell culture hinder their advancement in fundamental research and clinical/translational applications. As an emerging alternative technology, organ-on-a...
详细信息
During the coal seam drilling process, the drill string is subject to compressive deformation, compounded by unpredictable variations in formation hardness and borehole wall friction, leading to challenges in maintain...
详细信息
In consumer electronics, affective computing allows devices to recognize and respond to the emotional states of individuals, thereby enhancing personalization and satisfaction. Electroencephalogram (EEG)-based emotion...
详细信息
As China's steel production accounts for an increasing share of the world's output, the intelligent transformation of the steel industry is becoming increasingly urgent. To address issues such as low levels of...
详细信息
Integrating intermittent wind power into power systems results in low or zero inertia, threatening their frequency stability. To accommodate intermittent generations, the demand response(DR) is introduced, and air con...
详细信息
Integrating intermittent wind power into power systems results in low or zero inertia, threatening their frequency stability. To accommodate intermittent generations, the demand response(DR) is introduced, and air conditioning loads(ACs) account for an increasing proportion of all loads. The replacement of traditional generators with wind turbines and the ACs user uncertainties produce parameter uncertainties. This paper aims to construct an equivalent input disturbance(EID)-based load frequency control(LFC) strategy for the power system by considering wind power and ACs. First, an open-loop model is constructed for the LFC scheme with parameter uncertainties. Then, the parameter uncertainties and external disturbance are lumped into a fictitious disturbance, which is estimated using an EID *** incorporating the estimation of disturbance into the control input, the disturbance-rejection performance is achieved. Next, the Lyapunov theory is used to derive the two linear-matrix-inequality-based asymptotic stability criteria. A design algorithm is developed for the EID-based LFC scheme by exploiting an overall performance evaluation index. Finally, simulation results for the single-area and two-area LFC schemes show that, compared with the existing approaches, the method presented realizes the better disturbance rejection and higher robustness against parameter uncertainties, wind power fluctuation, and tie-line power ***, its robustness to time delays is verified.
In order to improve the safety protection performance of the rehabilitation robot,an active safety protection method is proposed in the rehabilitation *** oxyhemoglobin concentration information and RGB-D information ...
详细信息
In order to improve the safety protection performance of the rehabilitation robot,an active safety protection method is proposed in the rehabilitation *** oxyhemoglobin concentration information and RGB-D information are combined in this method,which aims to realize the comprehensive monitoring of the invasion target,the patient’s brain function movement state,and the joint angle in the rehabilitation *** main focus is to study the fusion method of the oxyhemoglobin concentration information and RGB-D information in the rehabilitation *** analysis of brain functional connectivity coefficient was used to distinguish the basic motion *** human skeleton recognition algorithm was used to realize the angle monitoring of the upper limb joint combined with the depth *** with speed and separation monitoring,the protection method of multi-information fusion is safer and more comprehensive for stroke *** building the active safety protection platform of the upper limb rehabilitation robot,the performance of the system in different safety states is tested,and the safety protection performance of the method in the upper limb rehabilitation scene is verified.
For safety reasons,in the automated dispensing medicines process,robots and humans cooperate to accomplish the task of drug sorting and *** this dynamic unstructured environment,such as a humanrobot collaboration scen...
详细信息
For safety reasons,in the automated dispensing medicines process,robots and humans cooperate to accomplish the task of drug sorting and *** this dynamic unstructured environment,such as a humanrobot collaboration scenario,the safety of human,robot,and equipment in the environment is *** this work,a practical and effective robot motion planning method is proposed for dynamic unstructured *** figure out the problems of blind zones of single depth sensor and dynamic obstacle avoidance,we first propose a method for establishing offline mapping and online fusion of multi-sensor depth images and 3D grids of the robot workspace,which is used to determine the occupation states of the 3D grids occluded by robots and obstacles and to conduct real-time estimation of the minimum distance between the robot and ***,based on the reactive control method,the attractive and repulsive forces are calculated and transformed into robot joint velocities to avoid obstacles in real ***,the robot’s dynamic obstacle avoidance ability is evaluated on an experimental platform with a UR5 robot and two KinectV2 RGB-D sensors,and the effectiveness of the proposed method is verified.
Urban underground comprehensive pipe galleries are gradually replacing traditional underground pipe galleries due to their advantages of strong safety, easy maintenance and monitoring. In order to monitor the power sy...
详细信息
暂无评论