To provide point-to-point output motions needed for automation, intermittent indexing tables have widely been employed in various industries. For ensuring adequate production volume and profit margins, such devices are usually required to accurately generate intermittent motions as rapidly as possible. As a result, both of the positioning time and residual vibration are major concerns with the design of a fast, intermittent indexing table. A procedure that can be used to synthesize motor motion curves with minimized positioning time for intermittent indexing tables at high speeds is presented. First, based on the measured dynamic characteristic of an indexing table, it is then simulated to derive the equations of motion. Subsequently, favorable parameters for defining asymmetrical motor motion commands by S-curves for the modeled indexing table with improved kinematic and dynamic performance are identified. To verify the accuracy and effectiveness of the proposed approach, numerical and experimental results are revealed and compared to those obtained by using the traditional method.

This research is mainly about autonomously navigation of an agricultural wheeled mobile robot in an unstructured outdoor setting. This project has four distinct phases defined as: (i) Navigation and control of a wheeled mobile robot for a point-to-point motion. (ii) Navigation and control of a wheeled mobile robot in following a given path (path following problem). (iii) Navigation and control of a mobile robot, keeping a constant proximity distance with the given paths or plant rows (proximity-following). (iv) Navigation of the mobile robot in rut following in farm fields. A rut is a long deep track formed by the repeated passage of wheeled vehicles in soft terrains such as mud, sand, and snow. To develop reliable navigation approaches to fulfill each part of this project, three main steps are accomplished: literature review, modeling and computer simulation of wheeled mobile robots, and actual experimental tests in outdoor settings. First, point-to-point motion planning of a mobile robot is studied; a fuzzy-logic based (FLB) approach is proposed for real-time autonomous path planning of the robot in unstructured environment. Simulation and experimental evaluations shows that FLB approach is able to cope with different dynamic and unforeseen situations by tuning a safety margin. Comparison of FLB results with vector field histogram (VFH) and preference-based fuzzy (PBF) approaches, reveals that FLB approach produces shorter and smoother paths toward the goal in almost all of the test cases examined. Then, a novel human-inspired method (HIM) is introduced. HIM is inspired by human behavior in navigation from one point to a specified goal point. A human-like reasoning ability about the situations to reach a predefined goal point while avoiding any static, moving and unforeseen obstacles are given to the robot by HIM. Comparison of HIM results with FLB suggests that HIM is more efficient and effective than FLB. Afterward, navigation strategies are built up for path following, rut following, and proximity-following control of a wheeled mobile robot in outdoor (farm) settings and off-road terrains. The proposed system is composed of different modules which are: sensor data analysis, obstacle detection, obstacle avoidance, goal seeking, and path tracking. The capabilities of the proposed navigation strategies are evaluated in variety of field experiments; the results show that the proposed approach is able to detect and follow rows of bushes robustly. This action is used for spraying plant rows in farm field. Finally, obstacle detection and obstacle avoidance modules are developed in navigation system. These modules enables the robot to detect holes or ground depressions (negative obstacles), that are inherent parts of farm settings, and also over ground level obstacles (positive obstacles) in real-time at a safe distance from the robot. Experimental tests are carried out on two mobile robots (PowerBot and Grizzly) in outdoor and real farm fields. Grizzly utilizes a 3D-laser range-finder to detect objects and perceive the environment, and a RTK-DGPS unit for localization. PowerBot uses sonar sensors and a laser range-finder for obstacle detection. The experiments demonstrate the capability of the proposed technique in successfully detecting and avoiding different types of obstacles both positive and negative in variety of scenarios.