Multi‐target detection and grasping control for humanoid robot NAO

Zhang, Lei; Zhang, Huayan; Yang, Hanting; Bian, Gui‐Bin; Wu, Wanqing

doi:10.1002/acs.3031

Cited by 13 publications

(10 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Such issues could be occlusion, background interference, and sudden move of the target. Besides, Zhang et al [48], proposed a robotic grasping method that uses the deep learning method you only look once for multi-target detection and the auxiliary signs to obtain target location. The method can control the movement of the robot and plan the grasping trajectory based on the visual feedback information.…”

Section: Related Workmentioning

confidence: 99%

Active object search using a pyramid approach to determine the next-best-view

Flores-Rodriguez

Trujillo-Romero

González-Barbosa

2022

IJRA

View full text Add to dashboard Cite

<p>The development of service robotics continues to arouse interest in the scientific community due to the complexity of the activities performed like interaction in human environments, identifying and manipulating objects, and even learning by themselves. This paper proposed to improve the perception of the environment by searching for objects in service robotics tasks. We present the development and implementation of an active object search method based on three main phases: Firstly, image pyramid segmentation to examine in detail the im- age features. Second step, object detection at each level of the pyramid through a local feature descriptor and a mutual information calculation. Finally, the next camera position selection through analyzing the object detections accumulation in the pyramid. To evaluate the implementation of the proposed method, we use a NAO robot in a familiar place for humans, such as an office or a home. Ordinary objects are part of our database with the premise that a robot must know them before looking for an object. The results in the experiments showed an acceptable performance in simulation and with a real platform.</p>

show abstract

Section: Related Workmentioning

confidence: 99%

Active object search using a pyramid approach to determine the next-best-view

Flores-Rodriguez

Trujillo-Romero

González-Barbosa

2022

IJRA

View full text Add to dashboard Cite

show abstract

“…The authors in [13] introduced a new approach to promote the exchange of objects between robots and humans. The authors in [23] propose, via the grasping for stacking network (GSN), a system that jointly learns the grasping and the stacking policies. This (GSN) enables a robotic arm to pick boxes from a table and put them on a platform properly.…”

Section: Grasp Taskmentioning

confidence: 99%

“…Object detection is one of the most important tasks in computer vision, which involves determining the location of certain objects in the image if they are present, as well as classifying those objects [23,24].…”

Section: Object Detectionmentioning

confidence: 99%

Vision-Based Robotic Arm Control Algorithm Using Deep Reinforcement Learning for Autonomous Objects Grasping

et al. 2021

View full text Add to dashboard Cite

While working side-by-side, humans and robots complete each other nowadays, and we may say that they work hand in hand. This study aims to evolve the grasping task by reaching the intended object based on deep reinforcement learning. Thereby, in this paper, we propose a deep deterministic policy gradient approach that can be applied to a numerous-degrees-of-freedom robotic arm towards autonomous objects grasping according to their classification and a given task. In this study, this approach is realized by a five-degrees-of-freedom robotic arm that reaches the targeted object using the inverse kinematics method. You Only Look Once v5 is employed for object detection, and backward projection is used to detect the three-dimensional position of the target. After computing the angles of the joints at the detected position by inverse kinematics, the robot’s arm is moved towards the target object’s emplacement thanks to the algorithm. Our approach provides a neural inverse kinematics solution that increases overall performance, and its simulation results reveal its advantages compared to the traditional one. The robot’s end grip joint can reach the targeted location by calculating the angle of every joint with an acceptable range of error. However, the accuracy of the angle and the posture are satisfied. Experiments reveal the performance of our proposal compared to the state-of-the-art approaches in vision-based grasp tasks. This is a new approach to grasp an object by referring to inverse kinematics. This method is not only easier than the standard one but is also more meaningful for multi-degrees of freedom robots.

show abstract

“…Assad-UZ-Zaman [7] et al developed a scheme to solve the inverse kinematics of the arm by using geometric method, and conducted remote simulation operation on the arm of NAO humanoid robot.Muller,J [8] et al proposed the grasping system of NAO humanoid robot composed of object detection device and grasping motion device. It can use the detection device to detect known small objects, and then use the grasping motion device to plan the arm trajectory and achieve the goal of grasping small objects.Zhai [9] et al researched the arm motion planning process of a robot for object capture.Zhang [10] took the complexity of multiple objects into account in the environment, put forward a multi-target detection and object-grabbing control method based on deep learning Yolo, and performed an experimental verification on the NAO robot.Zhu Tehao [11] proposed a human motion visual perception algorithm for humanoid robot, which improves the accuracy of human motion data captured by using Kinect as the visual input device.In order to improve the generalization of robot trajectory generation algorithm, Lin Limin [12] proposed a method based on spatiotemporal feature template(STFT) to generate robot arm trajectory.Balmik [13] proposed a teleoperation framework based on Kinect , which allows NAO humanoid robot to simulate recognized human actions using hidden Markov model (HMM).Wen [14] analyzed the topology of the upper arm of the NAO robot model and established the dynamic model. On this basis,the fuzzy PD controller was used to control and simulate the upper arm of NAO robot.…”

Section: Introductionmentioning

confidence: 99%

Fuzzy Proportional-Derivative Based Robot Arm Control for Object Transfer

Chen

Tan

et al. 2022

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

With the development of robot technology, humanoid robot and human-robot interaction are gradually playing a more and more important role in our society. At present, the arm control of humanoid intelligent robot has become one of the hot research issues in the field of robot. This paper proposes a fuzzy based NAO robot hand control method to reduce the vibration and increase position accuracy during object transferring between NAO robot and human. The 5-DOF kinematics model of NAO robot right arm is first established. Motion path of the NAO robot arm during human-robot interaction is planned and conducted in Matlab and reality. The impact of control period, joint speed and trajectory error on vibration is observed and discussed. Based on the discussion, the fuzzy-PD control method is proposed. Experiments result indicates that the proposed Fuzzy-PD control method can reduce vibration and trajectory error, which can benefit the smooth motion control process of the NAO robot arm receiving objects in the process of human-robot interaction.

show abstract

Multi‐target detection and grasping control for humanoid robot NAO

Cited by 13 publications

References 15 publications

Active object search using a pyramid approach to determine the next-best-view

Active object search using a pyramid approach to determine the next-best-view

Vision-Based Robotic Arm Control Algorithm Using Deep Reinforcement Learning for Autonomous Objects Grasping

Fuzzy Proportional-Derivative Based Robot Arm Control for Object Transfer

Contact Info

Product

Resources

About