Robotic grasps play an important role in the service and industrial fields, and the robotic arm can grasp the object properly depends on the accuracy of the grasping detection result. In order to predict grasping detection positions for known or unknown objects by a modular robotic system, a convolutional neural network(CNN) with the residual block is proposed, which can be used to generate accurate grasping detection for input images of the scene. The proposed model architecture was trained on the standard Cornell grasp dataset and evaluated on the test dataset. Moreover, it was evaluated on different types of household objects and cluttered multi-objects. On the Cornell grasp dataset, the accuracy of the model on image-wise splitting detection and object-wise splitting detection achieved 95.5％ and 93.6％, respectively. Further, the real detection time per image was 109 ms. The experimental results show that the model can quickly detect the grasping positions of a single object or multiple objects in image pixels in real time, and it keeps good stability and robustness.

WU Di;WU Nailong;SHI Hongrui

College of Information Science and Technology,Donghua University,Shanghai 201620,China

东华大学学报（英文版）

[1]WU Di;WU Nailong;SHI Hongrui-.Object Grasping Detection Based on Residual Convolutional Neural Network)[J].东华大学学报（英文版）,2022(04):345-352

grasps,cluttered

Object,Grasping,Detection,Based,Residual,Convolutional,Neural,Network,Robotic,play,important,role,service,industrial,fields,robotic,arm,can,properly,depends,accuracy,grasping,detection,In,order,predict,positions,unknown,objects,by,modular,system,convolutional,neural,network,residual,block,proposed,which,be,used,generate,accurate,input,images,scene,model,architecture,was,trained,standard,Cornell,dataset,evaluated,test,Moreover,different,types,household,On,wise,splitting,achieved,respectively,Further,real,ms,experimental,results,show,that,quickly,single,multiple,pixels,keeps,good,stability,robustness

0.506461