首页|Studies from Henan University of Science and Technology Provide New Data on Robo tics (Construction of Three-Dimensional Semantic Maps of Unstructured Lawn Scene s Based on Deep Learning)
Studies from Henan University of Science and Technology Provide New Data on Robo tics (Construction of Three-Dimensional Semantic Maps of Unstructured Lawn Scene s Based on Deep Learning)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-Investigators discuss new findings in robotics. According to news originating from Luoyang, People's Republic of China , by NewsRx correspondents, research stated, "Traditional automatic gardening pr uning robots generally employ electronic fences for the delineation of working b oundaries." Financial supporters for this research include Longmen Laboratory "trendy Indust ry Projects". Our news correspondents obtained a quote from the research from Henan University of Science and Technology: "In order to quickly determine the working area of a robot, we combined an improved DeepLabv3+ semantic segmentation model with a si multaneous localization and mapping (SLAM) system to construct a three-dimension al (3D) semantic map. To reduce the computational cost of its future deployment in resource-constrained mobile robots, we replaced the backbone network of DeepL abv3+, ResNet50, with MobileNetV2 to decrease the number of network parameters a nd improve recognition speed. In addition, we introduced an efficient channel at tention network attention mechanism to enhance the accuracy of the neural networ k, forming an improved Multiclass MobileNetV2 ECA DeepLabv3+ (MM-ED) network mod el. Through the integration of this model with the SLAM system, the entire frame work was able to generate a 3D semantic point cloud map of a lawn working area a nd convert it into octree and occupancy grid maps, providing technical support f or future autonomous robot operation and navigation. We created a lawn dataset c ontaining 7500 images, using our own annotated images as ground truth. This data set was employed for experimental purposes."
Henan University of Science and Technolo gyLuoyangPeople's Republic of ChinaAsiaEmerging TechnologiesMachine Le arningNano-robotRobotRobotics
NETL
NSTL
