China
Africa
Explore chapters and articles related to this topic
Robotics and Sensors: Environmental Applications
Published in John G. Webster, Halit Eren, Measurement, Instrumentation, and Sensors Handbook, 2017
Robots come in many different forms and sizes. Based on the method used for the locomotion of a robot, it can be classified as wheeled robot (including track robots) or legged robot. Legged robots may be two-legged humanoid or multilegged. Hybrid robots combine the aspects of wheeled and legged locomotion. Yet there are others types which cannot be classified as either wheeled or legged such as hovercrafts, underwater floating robots, and robots employing snakelike movements. A few different types of robots are shown in Figure 83.1.
Design of neural network–based PID controller for biped robot while ascending and descending the staircase
Published in Mangey Ram, J. Paulo Davim, Advanced Mathematical Techniques in Engineering Sciences, 2018
Ravi Kumar Mandava, Pandu R. Vundavilli
Compared to industrial manipulators, legged robots are having much more interaction with the ground and it is a tough job to control the robot in an effective manner. Further, the mechanism, structure and balancing of the two-legged robot is complex in nature when compared with other legged robots. Over the past few decades, people are working on the stability and controlling aspects of the biped robot on various terrains. Generating the stable gait for the biped robot while walking on various terrains is a difficult task and was taken up by many researchers. Presently, researchers are utilizing zero moment point (ZMP) [1] based control algorithms to control the gait of the two-legged robot. Some other researchers had tried to optimize the parameters of the ZMP-based controller after utilizing nontraditional algorithms [2,3]. It is important to note that the conventional PID controllers were widely deployed in both industrial as well as nonindustrial applications due to its ease, simple design, and cost effectiveness. Based on the demand for the usage of PID controller in various applications, the real-time tuning/adaption of the controller gains (i.e., Kp, Kd, and Ki) in an online manner is a challenging task. Moreover, the tuning methods such as Zeigler–Nicholas [4] and Cohen [5] methods were already proved that it is not possible to use them in highly nonlinear, uncertain, and coupled robotic applications. The advancements in the computational techniques motivated several researchers to use evolutionary and nature-inspired optimization algorithms, such as genetic algorithms (GAs) [6], particle swarm optimization (PSO) [7,8], ant colony optimization (ACO) [9], cuckoo search algorithm (CSA) [10] and bacterial forging optimization (BFO) [11], for tuning the gains of the conventional PID controller. The evolutionary and nature-inspired optimization algorithms provide a fixed set of optimal tuning parameters (i.e., gains). This fixed set of gains most of the time could not provide optimum performance when used in uncertain and nonlinear processes. Therefore, an efficient and effective online tuning mechanism is required for controlling the joints of the robot in a systematic manner. It is also important to mention that in order to improve the performance of the conventional PID controller, few researchers had tried with fuzzy logic technique for obtaining the optimal gains of the PID controller in an online manner. The PID controller was successfully implemented in various industrial applications, such as induction machine, ship plant [12–14], micro grid [15], hybrid electrical vehicle system [16] and AVR system [17], to control them in a more effective manner.
Generation of a continuous free gait for quadruped robot over rough terrains
Published in Advanced Robotics, 2019
Shuaishuai Zhang, Mingqu Fan, YiBin Li, Xuewen Rong, Ming Liu
Compared with the wheeled and tracked robot, the legged robot can travel on the rough terrains. Among the legged robots, the quadruped robots are more stable than biped robots and have less complicated mechanism [1]. Because of this, many researchers are attracted to pay attention to the research of the quadruped robot.