Learning of a Basketball Free Throw With a Flexible Link Robot

2021 ◽  
Author(s):  
Jannik Timke ◽  
Merlin Morlock ◽  
Daniel A. Duecker ◽  
Robert Seifried
Keyword(s):  
Real World ◽  
Control Method ◽  
Learning Control ◽  
The Finite Element Method ◽  
Flexible Link ◽  
Training Time ◽  
Free Throw ◽  
Rigid Link ◽  
Throwing Accuracy ◽  
Real World Learning

Abstract Object throwing is an efficient approach for overcoming the kinematic workspace limitations of robots in placement scenarios. Throwing of objects with rigid link robots has been widely studied in literature. Although using robots with spring-like flexible links can significantly increase the throwing distance, existing contributions are very rare. Therefore, we propose an efficient iterative learning control throwing algorithm and apply it to a flexible link robot. A simple rigid link throwing model is used to generate the motor motion. Errors caused by this simplification are corrected by a flexible link throwing model based on the finite element method. As representative scenario a basketball free throw is selected which requires high throwing accuracy. Here, we demonstrate that the controller can be efficiently pre-learned in simulations to reduce real-world training time. Experiments then validate that our learning control method achieves the required free throw accuracy within very few real-world learning iterations.

scholarly journals Task-space regulation of rigid-link electrically-driven robots with uncertain kinematics using neural networks

2021 ◽  
Vol 54 (1-2) ◽  
pp. 102-115
Author(s):  
Wenhui Si ◽  
Lingyan Zhao ◽  
Jianping Wei ◽  
Zhiguang Guan
Keyword(s):  
Neural Networks ◽  
Asymptotic Stability ◽  
Control Method ◽  
Joint Space ◽  
Robot Kinematics ◽  
Rbf Neural Networks ◽  
Task Space ◽  
Actuator Dynamics ◽  
Rigid Link ◽  
On Line

Extensive research efforts have been made to address the motion control of rigid-link electrically-driven (RLED) robots in literature. However, most existing results were designed in joint space and need to be converted to task space as more and more control tasks are defined in their operational space. In this work, the direct task-space regulation of RLED robots with uncertain kinematics is studied by using neural networks (NN) technique. Radial basis function (RBF) neural networks are used to estimate complicated and calibration heavy robot kinematics and dynamics. The NN weights are updated on-line through two adaptation laws without the necessity of off-line training. Compared with most existing NN-based robot control results, the novelty of the proposed method lies in that asymptotic stability of the overall system can be achieved instead of just uniformly ultimately bounded (UUB) stability. Moreover, the proposed control method can tolerate not only the actuator dynamics uncertainty but also the uncertainty in robot kinematics by adopting an adaptive Jacobian matrix. The asymptotic stability of the overall system is proven rigorously through Lyapunov analysis. Numerical studies have been carried out to verify efficiency of the proposed method.

scholarly journals How to train your robot with deep reinforcement learning: lessons we have learned

2021 ◽  
pp. 027836492098785
Author(s):  
Julian Ibarz ◽  
Jie Tan ◽  
Chelsea Finn ◽  
Mrinal Kalakrishnan ◽  
Peter Pastor ◽  
...  
Keyword(s):  
Machine Learning ◽  
Reinforcement Learning ◽  
Case Studies ◽  
Real World ◽  
Review Article ◽  
The Real ◽  
Complex Skills ◽  
Real World Learning ◽  
Level Sensor ◽  
Embodied Agent

Deep reinforcement learning (RL) has emerged as a promising approach for autonomously acquiring complex behaviors from low-level sensor observations. Although a large portion of deep RL research has focused on applications in video games and simulated control, which does not connect with the constraints of learning in real environments, deep RL has also demonstrated promise in enabling physical robots to learn complex skills in the real world. At the same time, real-world robotics provides an appealing domain for evaluating such algorithms, as it connects directly to how humans learn: as an embodied agent in the real world. Learning to perceive and move in the real world presents numerous challenges, some of which are easier to address than others, and some of which are often not considered in RL research that focuses only on simulated domains. In this review article, we present a number of case studies involving robotic deep RL. Building off of these case studies, we discuss commonly perceived challenges in deep RL and how they have been addressed in these works. We also provide an overview of other outstanding challenges, many of which are unique to the real-world robotics setting and are not often the focus of mainstream RL research. Our goal is to provide a resource both for roboticists and machine learning researchers who are interested in furthering the progress of deep RL in the real world.

Continuous reinforcement learning based ramp jump control for single-track two-wheeled robots

2021 ◽  
pp. 014233122110378
Author(s):  
Qingyuan Zheng ◽  
Duo Wang ◽  
Zhang Chen ◽  
Yiyong Sun ◽  
Bin Liang
Keyword(s):  
Reinforcement Learning ◽  
Energy Savings ◽  
Control Method ◽  
Learning Control ◽  
Action Space ◽  
Gradient Algorithm ◽  
Single Track ◽  
Wheeled Robots ◽  
Reward Function ◽  
Wheeled Robot

Single-track two-wheeled robots have become an important research topic in recent years, owing to their simple structure, energy savings and ability to run on narrow roads. However, the ramp jump remains a challenging task. In this study, we propose to realize a single-track two-wheeled robot ramp jump. We present a control method that employs continuous action reinforcement learning techniques for single-track two-wheeled robot control. We design a novel reward function for reinforcement learning, optimize the dimensions of the action space, and enable training under the deep deterministic policy gradient algorithm. Finally, we validate the control method through simulation experiments and successfully realize the single-track two-wheeled robot ramp jump task. Simulation results validate that the control method is effective and has several advantages over high-dimension action space control, reinforcement learning control of sparse reward function and discrete action reinforcement learning control.

scholarly journals The Validation of Digital Analysis Tool-Assisted Real-World Inquiry (Digita-RI) as a Modification of the Inquiry-Based Learning Model in the Digital Age

2021 ◽  
Vol 10 (3) ◽  
pp. 387-399
Author(s):  
S. Nurohman ◽  
W. Sunarno ◽  
S. Sarwanto ◽  
S. Yamtinah
Keyword(s):  
Real World ◽  
Group Discussion ◽  
Scientific Work ◽  
Learning Model ◽  
Analysis Tool ◽  
Digital Analysis ◽  
Inquiry Based Learning ◽  
Real World Learning ◽  
Feasibility Test ◽  
Scientific Facts

Inquiry-based learning has been tested to improve conceptual understanding, reduce misconceptions, and provide students with experiences in scientific work. However, in its implementation, inquiry-based learning is often faced with scientific facts from the real world with data which hard to analyze using traditional methods. Therefore, a breakthrough is needed to overcome the weaknesses of inquiry-based learning by integrating digital analysis tools and the concept of real-world learning. This integration produces a new learning model, the Digital Analysis Tool-Assisted Real-World Inquiry (Digita-RI). This study aims to test the feasibility and practicality of the Digita-RI learning model. This Research and Development (R&D) use the steps proposed by Barg and Gall. The feasibility test of the Digita-RI model was carried out through the Focus Group Discussion (FGD) method and the assessment of the Digita-RI model book involving seven experts. The practicality test was carried out through the Think Aloud Protocol (TAP), and the assessment of the Digita-RI model guidebook involved five practitioner lecturers and six students. The results of expert, practitioner, and user assessments were analyzed using the Aiken coefficient (Aiken’s V). The results showed that Digita-RI is a feasible and practical learning model. Therefore, it can be concluded that Digita-RI has the feasibility and practicality to be used in science learning in the classroom.

scholarly journals Undergraduates Writing Publishable Cases: The Trials, The Tribulations, And The Real-World Learning

2011 ◽  
Vol 4 (12) ◽  
Author(s):  
Jason Smith ◽  
Josh Edwards ◽  
Patricia C. Kelley
Keyword(s):  
Case Studies ◽  
Real World ◽  
The Real ◽  
Real World Learning ◽  
Case Writing

If given the chance, undergraduates have the ability to write excellent case studies worthy of being published.  This essay describes the benefits, challenges, and process of undergraduate case writing. 

Sign in / Sign up

Export Citation Format

Share Document