Design and Development of Intelligent Controller for 7-DoF Anthropomorphic Arm to Generate Human-Like Arm Movements.

In the modern age, robotics is an attractive field for many researchers as robots are involved in many aspects of everyday life due to the satisfaction and solutions that they provide in various daily difficulties. The problem is that human arm movements are complex. During the movement, arm states

Project Title

Design and Development of Intelligent Controller for 7-DoF Anthropomorphic Arm to Generate Human-Like Arm Movements.

Project Area of Specialization

Robotics

Project Summary

In the modern age, robotics is an attractive field for many researchers as robots are involved in many aspects of everyday life due to the satisfaction and solutions that they provide in various daily difficulties. The problem is that human arm movements are complex. During the movement, arm states are changing constantly. Different numbers and combinations of joints form different arm states. Human arms can perform different and complex tasks by coordination and cooperation of different joints. The Generation of Human-Like Arm Movements for Anthropomorphic Arm and the solution of inverse kinematics for object reach task is a challenging problem that seems more attractive to researchers as years pass by. In our implementation (project),  the goal is not only to find one random (of the infinite solutions) of these problems, but to determine the one that minimizes both the position error between the actual and desired position of the end-effector of the robotic arm and the Generation of Human-Like Arm Movements for Anthropomorphic Arm. These problems can be solved by using human-like motion planning strategy. The strategy consists of three parts: movement primitives, Bayesian network (BN), and a novel coupling neural network (CPNN). The movement primitives will use to decouple the human arm movements. The classification of arm movements improves the accuracy of human-like movements. The motion-decision algorithm based on BN will able to predict occurrence probabilities of the motions and choose appropriate mode of motion. Then, a novel CPNN will proposed to solve the inverse kinematics problems of humanoid robotic arms. The strategy was used by researchers in and resolved the problem of inverse kinematic and achieved human like arm motion with average accuracy of 96.71 (%) for 5-Dof Anthropomorphic Arm. So, In this project (FYP), efforts will be exerted to resolve the problem of inverse kinematic and generation of human like arm motion for 7-Dof Anthropomorphic Arm with average high accuracy.

Project Objectives

1. Design and Development of Bayesian Network for motion prediction.
2. Development of a Coupling Neural Network model for 7-DoF movement control.
3. Mechanical model development of 7-DoF Arm.
4. Resolution of redundancy.
5. Minimize singularities & non-linearities.

Project Implementation Method

The strategy consists of three parts: movement primitives, Bayesian network (BN), and a novel coupling neural network (CPNN). The movement primitives are used to decouple the human arm movements. The classification of arm movements improves the accuracy of human-like movements. The motion-decision algorithm based on BN is able to predict occurrence probabilities of the motions and choose appropriate mode of motion. Then, a novel CPNN is proposed to solve the inverse kinematics problems of anthropomorphic arms.

The structure of method to generate human-like movement for object reach task.

The flowchart of the motion-decision algorithm for object reach task.

Benefits of the Project

Humanoid Robots are being used in a lot of branches of our life such as home maintenance robot, Robots as explorers, humanoid cooking robot etc. Many spaceships are operated by robots. It is sometimes a very dangerous job for a human. Robots do this for us. Recently, NASA sent a robot named “curiosity” to explore the Mars. It is not possible for a human There are many programmable robots which can cook food according to your choice. You just have to set the quantity of the ingredients of the food. The rest will be done by the robot. Many robots are now introduced which can copy you. All you need is just do the cooking in front of the robot for once. The movement of your body is recorded by the camera. From then, the robot will copy your actions to make that food for you. This kind of humanoid robotic cooking assistants can be introduced in many hotels, houses and offices of Pakistan. As for humanoid robot to work smoothly and efficiently its arms should generate human like movement on which we are doing work. After the completion of this project if government support this project we can manufacture them in Pakistan at a very low cost and its sale at local or its export will become a great achievement for Pakistan. And consequently it be beneficial for the GDP of Pakistan.

Technical Details of Final Deliverable

A humanoid robotic arm equipped with the replicated abilities of humans will be available as a deliverable. The developed robotic Anthropomorphic Arm will be fully functional to be utilized for any object reach tasks with high precision.

Final Deliverable of the Project

Hardware System

Core Industry

Others

Other Industries

Core Technology

Robotics

Other Technologies

Artificial Intelligence(AI)

Sustainable Development Goals

Required Resources

If you need this project, please contact me on contact@adikhanofficial.com