Current Issue : October-December Volume : 2022 Issue Number : 4 Articles : 5 Articles
The biped robot adopts the human movement mode. Compared with other movement modes, the gait has good flexibility and adaptability. It is very important in the research of robotics, so it has become a hot spot of robotics research. This article aims to study the application of neural networks in the stability of biped robots and the embedded control of walking mode. A method of establishing precise mathematical modeling and stability analysis is proposed. Based on this model, the motion characteristics of the biped robot’s walking mode and the local stability of joints are studied, and the motion mode of passive walking under the control of the neural network is deeply analyzed, using a neural network to control the stability of biped robot motion and adopting the research method of the embedded control system in walking mode. Essentially, the output value of the physical network is used to judge whether the robot is in a stable position so as to perform appropriate actions and control the robot’s stability of walking. The experimental results show that the biped robot can detect movement and overcome obstacles through related networks and embedded control systems. Through the control of the embedded system, the errors of each joint of the biped robot on flat ground, stairs, and obstacles are greatly reduced. The most obvious reduction of the deviation is that the ankle joint decreases from 2.5 to 0.07 when rotating, and the knee joint angle deviation is reduced from 3.8 to 2, which greatly improves the stability of the biped robot’s walking mode....
Machine learning algorithms are effective in realizing the programming of robots that behave autonomously for various tasks. For example, reinforcement learning (RL) does not require supervision or data sets; the RL agent explores solutions by itself. However, RL requires a long learning time, particularly for actual robot learning situations. Transfer learning (TL) in RL has been proposed to address this limitation. TL realizes fast adaptation and decreases the problemsolving time by utilizing the knowledge of the policy, value function, and Q-function from RL. Taylor proposed TL using inter-task mapping that defines the correspondence between the state and action between the source and target domains. Inter-task mapping is defined based on human intuition and experience; therefore, the effect of TL may not be obtained. The difference in robot shapes for TL is similar to the cognition in the modification of human body composition, and automatic inter-task mapping can be performed by referring to the body representation that is assumed to be stored in the human brain. In this paper, body calibration is proposed, which refers to the physical expression in the human brain. It realizes automatic inter-task mapping by acquiring data modeled on a body diagram that illustrates human body composition and posture. The proposed method is evaluated in a TL situation from a computer simulation of RL to actual robot control with a multi-legged robot....
This paper presents a novel design and development of a low-cost and multi-touch sensor based on capacitive variations. This new sensor is very flexible and easy to fabricate, making it an appropriate choice for soft robot applications. Materials (conductive ink, silicone, and control boards) used in this sensor are inexpensive and easily found in the market. The proposed sensor is made of a wafer of different layers, silicone layers with electrically conductive ink, and a pressuresensitive conductive paper sheet. Previous approaches like e-skin can measure the contact point or pressure of conductive objects like the human body or finger, while the proposed design enables the sensor to detect the object’s contact point and the applied force without considering the material conductivity of the object. The sensor can detect five multi-touch points at the same time. A neural network architecture is used to calibrate the applied force with acceptable accuracy in the presence of noise, variation in gains, and non-linearity. The force measured in real time by a commercial precise force sensor (ATI) is mapped with the produced voltage obtained by changing the layers’ capacitance between two electrode layers. Finally, the soft robot gripper embedding the suggested tactile sensor is utilized to grasp an object with position and force feedback signals....
A robot path planning algorithm based on reinforcement learning is proposed. The algorithm discretizes the information of obstacles around the mobile robot and the direction information of target points obtained by LiDAR into finite states, then reasonably designs the number of environment model and state space, and designs a continuous reward function, so that each action of the robot can be rewarded accordingly, which improves the algorithm and improves the training efficiency of the algorithm. Finally, the agent training simulation environment is built on gazebo, and the training results verify the effectiveness of the algorithm. At the same time, the navigation experiment is carried out on the actual robot. The experimental results show that the algorithm can also complete the navigation task in the real environment....
The wearable power-assisted robot is a typical auxiliary rehabilitation robot. It is an exoskeleton power-assisted device that helps people to expand their lower limb movement capabilities. Its basic principle is to obtain the motion intention information of the human body through the perception system. Control the DC servo motor installed at the hip joint and the knee joint to drive the movement of the link, so as to achieve the purpose of providing assistance to the human body. In order to improve the dynamic response frequency of the wearable robotic perception system, a sensor signal based on time series analysis is proposed. The online prediction algorithm, which can perform single-step or multistep prediction under the premise of ensuring certain accuracy, can multiply the dynamic response frequency of the wearable-assisted robot sensing system to ensure the real-time performance of the whole system. In order to realize the sensor signal prediction algorithm, we design the corresponding software and hardware system to realize the prediction algorithm. The whole sensor signal prediction algorithm implementation system can be divided into two parts: lower computer and upper computer. The lower computer includes amplification circuit, signal conditioning circuit, and acquisition. The signal processing software part of the circuit and the corresponding MCU and the upper computer mainly include the data acquisition and prediction algorithm implementation, and the upper computer adopts the mixed programming technology of Vc++ and MATLAB to complete the software part of the upper computer. Aiming at the control part of the wearable robotic sensing system, the first generation of DC servo motor embedded motion controller is designed. The motion controller adopts the design concept of embedded motion controller, which has small size, is light weight, and has good expandability. And the motion controller can communicate and debug with the host computer through the serial port, which lays a foundation for the design of the entire embedded control system....
Loading....