17 February 2020 in IEEE/ASME Transactions on Mechatronics
Due to the features of soft material such as softness, deformability and adaptability to constrained surroundings, soft manipulators can collide with external objects in some extent. To take the best advantage of this merit, this paper presents a method of visual servo considering collisions with obstacles for a soft manipulator. First of all, a collision detection algorithm is proposed using shape feedbacks obtained by a sensor network, on the basis of the assumption that the manipulator can be considered as a combination of a static segment and a dynamic segment. Then, a contact localization algorithm is proposed to pinpoint the position where the collision occurs by minimizing the differences between the theoretical and actual shapes. An obstacle get-away algorithm, which is achieved by driving the calculated contact point to a desired position, is presented to avoid the obstacle. Along with the visual servo, a controller taking the obstacle avoidance into account is proposed to perform the hybrid task, with its stability validated via Lyaponov criterion. Finally, simulations and physical experiments are carried out to evaluate the algorithms. It is shown that the pixel error can converge to zero and the distance from the obstacle with respect to the contact point increases obviously in a short time.
13 October 2023 in IEEE Transactions on Industrial Electronics
Sport and game industry has grown rapidly in recent years due to the application of novel sensors and algorithms for quantitative analysis. For example, flying speed and spin estimation is essential to help players to improve their skills in table tennis. However, the spin estimation for a table tennis ball is challenging, as it is difficult to observe using cameras and model the aerodynamics of ball flight with spin. This article proposes a generalized aerodynamic model with variable aerodynamic coefficients to accurately represent the flying state of a table tennis ball. Analytical solutions for the aerodynamic coefficients and the acceleration due to the Magnus force are also developed for accurate ball spin estimation using pre- and postrebounding flight trajectories. The experimental results showed that compared to current state-of-the-art methods, the proposed method has achieved the best performance in angular velocity magnitude estimation for topspinning and backspinning balls. It also achieved an error of below 10° in angular velocity amplitude estimation. Using the proposed spin estimation method, our table tennis robot could strike balls with either topspin or backspin with a high success rate of up to 84.6 $%$ . Besides, the experimental results also demonstrated the potential of the proposed method in the area of table tennis training and sports-broadcasting.
6 November 2023 in IEEE/ASME Transactions on Mechatronics
Table-tennis robots still have a long way to defeat human players because of the rapid decrease in control accuracy when the ball flying speed and spin velocity increase. In this case, the challenge is to control the robot accurately within a limited time. To overcome this challenge, this article introduces a receding horizon optimization method to solve the accuracy problem of spin estimation. With the help of accurate spin information, a single-parameter-optimization-based stroke approach is developed for the robot to guarantee accuracy and efficiency simultaneously. In terms of accuracy, the proposed method achieves the best performance compared with other spin estimation algorithms, and it also achieves a stroke error of 25.13 cm and 23.93 cm for topspinning and backspinning balls, with the passing-net height error being 12.27 cm and 10.57 cm, respectively. Due to the high efficiency of the method proposed, our robot can hit a ball that has a speed of up to 10 m/s, which is a new record for the maximum ball speed that a table-tennis robot can handle to the best of our knowledge. Using the method proposed in this article achieves success rates of 90.0 $%$ and 85.0 $%$ for striking topspin and backspin balls, surpassing the respective success rates of 40.0 $%$ and 13.1 $%$ obtained by using the previous method.
4 October 2024 in Neural Processing Letters
Deep reinforcement learning has significantly advanced robot manipulations by providing an alternative solution for designing control strategies using raw images as direct inputs. While images offer additional environmental information, the end-to-end policy training manner (from image to action) requires simultaneous representation and task learning by the agent. This often necessitates a substantial number of interaction samples to achieve satisfactory policy performance. Previous works has attempted to address this challenge by learning a visual representation model that encodes the entire image into a low-dimensional vector before the policy training. However, since this vector contains both robot and object information, it inevitably introduces coupling within the state, which can mislead the policy training process. In this study, a novel method called Reinforcement Learning with Decoupled State Representation is proposed to effectively decouple robot and object information within the state representation. Experimental results demonstrate that the proposed method exhibits faster learning speed and achieves superior performance compared to previous methods across various robot manipulation tasks. Moreover, with only 3096 offline images, the proposed method successfully applies to real-world robot pushing tasks, which demonstrates its high practicability.
26 March 2024 in IEEE Transactions on Instrumentation and Measurement
In table tennis, developing a precise ball-racket rebound model is crucial for predicting the trajectory and spin of a ball after it hits the racket, which is instrumental in racket design and enhancing the capabilities of table tennis robots. To this end, accuracy and computational efficiency are two challenges to overcome, which has not been perfectly handled using existing methods such as finite element and simplified rigid-body models. This paper introduces a new model that calculates ball-racket rebounds in two orthogonal directions. Vertically, the collision dynamics are analyzed with the Kelvin-Voigt model, revealing that the contact duration is independent of the incoming ball’s velocity. Horizontally, we establish that the restitution coefficient varies as a function of the incident velocity, based on a continuous contact force model and momentum conservation principles. High-speed camera data corroborate these findings and confirm the model’s efficacy across diverse conditions. Compared to an established representative model, our method not only maintains high computational efficiency but also improves the accuracy of predicting the ball’s linear and angular velocities by an average of 42.72% and 33.77%, respectively, as evidenced by our experimental data.