Taqiaden Alshameri
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Taqiaden Alshameri has a total of 13 co-authors. Here are the top co-authors by number of shared publications:
Abdullah Alqadhi
School of Mechanical Engineering, Hefei University of Technology, Hefei, China
2 shared publications- Yude Dong
School of Mechanical Engineering, Hefei University of Technology, Hefei, China
2 shared publications Ahmed Maqram
1 shared publication- Daheng Li
State Key Laboratory of Multimodal Artificial Intelligence Systems, Institute of Automation, Chinese Academy of Sciences, Beijing, China
1 shared publication - Haonan Duan
State Key Laboratory of Multimodal Artificial Intelligence Systems, Institute of Automation, Chinese Academy of Sciences, Beijing, China
1 shared publication Heng Ding
1 shared publication
Latest Publications
- 19 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 10 (1) , 366-373
GraspAgent 1.0: Adversarial Continual Dexterous Grasp Learning
Grasp is at the core of robotic manipulation tasks. Nonetheless, most 6-DOF methods resort to a one-time setup via intensive analytics and targeting a predetermined domain. On the other hand, learning and adapting in real environments is of great promise to robotics yet challenging. In this context, this letter presents a grasp learning agent embodied in a dual-arm robot with a gripper and suction ends. Theoretically, GraspAgent is equipped with two adversarial components with the capacity for continual improvement. The first component aims to learn the synthesis of high dexterity grasps using a novel Grasp Awareness Generative Adversarial Network . Thanks to the designed set of contrastive objectives, GA-GAN yields a high rate of feasible grasps for clutter scenes. The second component is an Adversarial Experience Replay . AER promotes certain attributes during the regular rounds of training. For instance, a quality network is forced to adapt the maturity of the sampler by highlighting regions where the sampler performs well. Further, the quality network of each arm is encouraged to learn different grasp primitives, which reflects on better learning capacity. Finally, Experiments on a Yumi robot reveal a final average success rate of 93% after exploiting 17k+ feedback data. Moreover, We show that grasp sampling with GA-GAN surpasses a set of recent baseline methods.
- 19 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 10 (1) , 366-373
GraspAgent 1.0: Adversarial Continual Dexterous Grasp Learning
Grasp is at the core of robotic manipulation tasks. Nonetheless, most 6-DOF methods resort to a one-time setup via intensive analytics and targeting a predetermined domain. On the other hand, learning and adapting in real environments is of great promise to robotics yet challenging. In this context, this letter presents a grasp learning agent embodied in a dual-arm robot with a gripper and suction ends. Theoretically, GraspAgent is equipped with two adversarial components with the capacity for continual improvement. The first component aims to learn the synthesis of high dexterity grasps using a novel Grasp Awareness Generative Adversarial Network . Thanks to the designed set of contrastive objectives, GA-GAN yields a high rate of feasible grasps for clutter scenes. The second component is an Adversarial Experience Replay . AER promotes certain attributes during the regular rounds of training. For instance, a quality network is forced to adapt the maturity of the sampler by highlighting regions where the sampler performs well. Further, the quality network of each arm is encouraged to learn different grasp primitives, which reflects on better learning capacity. Finally, Experiments on a Yumi robot reveal a final average success rate of 93% after exploiting 17k+ feedback data. Moreover, We show that grasp sampling with GA-GAN surpasses a set of recent baseline methods.
- 28 January 2022
- Published by Springer Nature in The International Journal of Advanced Manufacturing Technology
- Vol. 119 (11-12) , 8003-8023
Declining neighborhood simulated annealing algorithm to robust fixture synthesis in a point-set domain
- 28 January 2022
- Published by Springer Nature in The International Journal of Advanced Manufacturing Technology
- Vol. 119 (11-12) , 8003-8023
Declining neighborhood simulated annealing algorithm to robust fixture synthesis in a point-set domain
- Open Access
- 14 July 2021
- Published by Springer Nature in Research Square
Declining Neighborhood Simulated Annealing Algorithm to Robust Fixture Synthesis in a Point-Set Domain
Fixture synthesis addresses the problem of fixture-elements placement on the workpiece surfaces. This article presents a novel variant of the Simulated Annealing (SA) algorithm called Declining Neighborhood Simulated Annealing (DNSA) specifically developed for the problem of fixture synthesis. The objective is to minimize measurement errors in the machined features induced by the misalignment at workpiece-locator contact points. The algorithm systematically evaluates different fixture layouts to reach a sufficient approximation of the global optimum robust layout. For each iteration, a set of previously accepted candidates are exploited to predict the next move. Throughout the progress of the algorithm, the search space is reduced and the new candidates are designated according to a declining Probability Density Function (PDF). To assure best performance, the DNSA parameters are configured using the Technique for Order Preference by Similarity to Ideal Solution (TOPOSIS). Moreover, the parameters are set to auto-adapt the complexity of a given input based on a Shanon entropy index. The optimization process is carried out automatically in the Computer-Aided Design (CAD) environment NX; a computer code was developed for this purpose using the Application Programming Interface (API) NXOpen. Benchmark examples from industrial partner and literature demonstrate satisfactory results.
- 14 July 2021
- Published by Springer Nature in Research Square
Open AccessDeclining Neighborhood Simulated Annealing Algorithm to Robust Fixture Synthesis in a Point-Set Domain
Fixture synthesis addresses the problem of fixture-elements placement on the workpiece surfaces. This article presents a novel variant of the Simulated Annealing (SA) algorithm called Declining Neighborhood Simulated Annealing (DNSA) specifically developed for the problem of fixture synthesis. The objective is to minimize measurement errors in the machined features induced by the misalignment at workpiece-locator contact points. The algorithm systematically evaluates different fixture layouts to reach a sufficient approximation of the global optimum robust layout. For each iteration, a set of previously accepted candidates are exploited to predict the next move. Throughout the progress of the algorithm, the search space is reduced and the new candidates are designated according to a declining Probability Density Function (PDF). To assure best performance, the DNSA parameters are configured using the Technique for Order Preference by Similarity to Ideal Solution (TOPOSIS). Moreover, the parameters are set to auto-adapt the complexity of a given input based on a Shanon entropy index. The optimization process is carried out automatically in the Computer-Aided Design (CAD) environment NX; a computer code was developed for this purpose using the Application Programming Interface (API) NXOpen. Benchmark examples from industrial partner and literature demonstrate satisfactory results.