Kalibracija senzora robota neuronskom mrežom i optimizacijom roja čestica poboljšana križanjem i mutacijom

• Main Authors: Ge, Dong-Yuan; Department of Mechanical and Energy Engineering, Shaoyang University, Shaoyang 422004, Hunan, China; School of Mechanical and Automotive Engineering, South China University of Technology, Guangzhou, 510640, Guangdong, China; [e-mail], Yao, Xi-Fan; School of Mechanical and Automotive Engineering, South China University of Technology, Guangzhou, 510640, Guangdong China, Yao, Qing-He; Department of Mechanical Engineering, Faculty of Engineering, Kyushu University, 744, Motooka, Nishi-ku, Fukuoka 819-0395, Japan, Jin, Hong; School of Mechanical and Automotive Engineering, South China University of Technology, Guangzhou, 510640, Guangdong China
• Format: Text
• Language: Croatian; English
• Published: Faculty of Mechanical Engineering in Slavonski Brod, Faculty of Electrical Engineering in Osijek, Faculty of Civil Engineering in Osijek; [e-mail] 2014
• Subjects: kalibracija senzora robota; lateralni pravac; longitudinalni pravac; neuronska mreža s rotacijskom matricom; optimalizacija roja čestica križanjem i mutacijom; crossover and mutation particle swarm optimization; lateral direction; longitudinal direction; neural network with rotational matrix; robot sensor calibration; Kad’; Roja; sami
• Online Access: http://hrcak.srce.hr/129051; http://hrcak.srce.hr/file/190625; http://hrcak.srce.hr/file/190626

U cilju određivanja položaja i orijentacije nekog predmeta u zglobu za robot, treba procijeniti odnos transformacije sustava ruka-oko, što se opisuje kao rotacijska matrica i vektor translacije. Predlaže se novi pristup koji integrira neuronsku mrežu i algoritam optimaizacije roja čestica s operacijom križanja i mutacije za kalibraciju osjećaja robota. Najprije se strukturira neuronska mreža s matricom rotacijske težine gdje su težine elementi rotacijskog dijela homogenog prijenosa sustava ruka-oko. Tada se algoritam optimalizacije roja čestica integrira u program rješavanja, gdje se faktori težine inercije i vjerojatnosti mutacije sami podešavaju prema putanji gibanja čestica u longitudinalnom pravcu i lateralnom pravcu. Kad je zadovoljen kriterij terminacije, rotaciona matrica se dobiva iz nepromjenljivih težina neuronske mreže. Tada se rješava vektor translacije i postiže se položaj i orijentacija slike s kamere u odnosu na sliku sa zgloba. Predloženi pristup pruža novu šemu za kalibraciju robota tehnikom samo-adaptacije, što garantira ortogonalnost riješenih rotacijskih komponenti homogenog transforma. In order to determine the position and orientation of an object in the wrist frame for robot, transform relation of hand-eye system should be estimated, which is described as rotational matrix and translational vector. A new approach integrating neural network and particle swarm optimization algorithm with crossover and mutation operation for robot sense calibration is proposed. First the neural network with rotational weight matrix is structured, where the weights are the elements of rotational part of homogeneous transform of the hand-eye system. Then the particle swarm optimization algorithm is integrated into the solving program, where the inertia weight factor and mutation probability are tuned self-adaptively according to the motion trajectory of particles in longitudinal direction and lateral direction. When the termination criterion is satisfied, the rotational matrix is obtained from the neural network’s stable weights. Then the translational vector is solved, so the position and orientation of camera frame with respect to wrist frame is achieved. The proposed approach provides a new scheme for robot sense calibration with self-adaptive technique, which guarantees the orthogonality of solved rotational components of the homogeneous transform.