China
Africa
Explore chapters and articles related to this topic
Positioning Methods
Published in Basudeb Bhatta, Global Navigation Satellite Systems, 2021
Dead-reckoning is the process of calculating one’s current position by using their previous positions—a technique of determining position by computing distance travelled on a given course (direction). Distance travelled is determined by multiplying speed by elapsed time. The principle of a dead-reckoning system is the relative position fixing method, which requires knowledge of the location of a vehicle and its subsequent speed and direction (for example, the last position and velocity determination before GNSS signal interruption) in order to calculate its present position (Bowditch 1995). A typical dead-reckoning system therefore comprises distance and heading (direction) sensors. Such a system can only give the 2D (horizontal) position of a vehicle (although more sophisticated dead-reckoning systems may include altitude sensors or inclinometers which can provide the 3D position of the vehicle). However, because of unfavourable error accumulation (a small error in heading grows over time into a large error in position), frequent calibration is required. It is in this context that GNSS is integrated with dead-reckoning systems. That is, the dead-reckoning sensors provide information on relative position (relative to a starting location), but GNSS receiver position measurements (x, y, z) are used to determine the dead-reckoning sensor errors, which may be fed back into the navigation computer.
Transferring Human Navigational Skills to Smart Wheelchair
Published in Yangsheng Xu, Ka Keung C. Lee, Human Behavior Learning and Transfer, 2005
However, in practice, the environments in which mobile robots operate are usually modeled in highly complex forms, such as geometric or image representations, and as a result autonomous navigation and localization can be difficult. The difficulties are exacerbated for practical robots with limited onboard computational resources and complex planning algorithms, since this paradigm of environmental modeling requires enormous computational power. In fact, environment modeling and localization can be considered as dual problems, because in order to localize a mobile robot many localization systems require a world model to match observed environment characteristics with the modeled ones. This is also true because in order to build a world model, most systems require precise localization of the robot [95], [137]. The most significant types of environment representations are cell-decomposition models, geometrical models, and topological models. Typically, modeling consists of several location sensing techniques, such as scene analysis, triangulation, proximity, and dead reckoning. The first technique, scene analysis, refers to the detection of scene features for inferring the objection location [49], [10]. The second, triangulation, refers to the use of the geometric properties of triangles to compute object location. The third, proximity, refers to the detection of an object when it is near to a known location by taking advantage of the limited range of a physical phenomenon. The fourth, dead reckoning, refers to incremental positioning methods such as odometry and inertial navigation systems.
Computerized Food Warehouse Automation
Published in Gauri S. Mittal, Computerized Control Systems in the Food Industry, 2018
J. Pemberton Cyrus, Lino R. Correia
An automated guided vehicle system (AGVS) is a system of autonomous driverless vehicles with programmable routes. The vehicles usually follow either a buried cable emitting a radiofrequency (RF) signal, or a painted line on the floor. Some newer AGVs are guided by bar-coded station tags or use “dead reckoning”. In the dead reckoning systems, sometimes called self-guided vehicles (SGVs), the vehicle maintains a record of its location based on calculations of distance traveled and turns executed. An AGVS is expensive, but it has route flexibility and low labor cost.
A new architecture for simultaneous localization and mapping: an application of a planetary rover
Published in Enterprise Information Systems, 2021
Kuo-Kun Tseng, Jun Li, Yachin Chang, K.L. Yung, C.Y. Chan, Chih-Yu Hsu
Currently, well-rounded navigation technologies are Dead reckoning, visual navigation, radio navigation. Dead reckoning often relies on the odometer or another unit for navigation, but the fatal drawback of dead navigating is the existence of cumulative errors increase as time goes on. Therefore, this location method isn’t suitable for long-term Lunar exploration mission. One characteristic of the Moon is that it has no atmosphere. The Moon has no atmosphere. Ultrasonic sensors are useless. The magnetic field of the Moon is very weak in comparison to that of the Earth. A magnetic compass would not work on the Moon because the magnetic field of the Moon is very weak in comparison to that of the Earth. The magnetic compass for navigation is useless on the moon. The GPS system operates using 50 or so satellites in orbit around the Earth. They would not be useful on the Moon even though their signals were possibly received. The Earth’s GPS positioning system cannot provide lunar rover navigation services. Thus, the most feasible method for Lunar rover locating is visual navigation. Based on our proposed MVMSLAM, we propose a lunar rover locating module, as shown in Figure 10.
The automated driver as a new road user
Published in Transport Reviews, 2021
Ane Dalsnes Storsæter, Kelly Pitera, Edward D. McCormack
Accelerometers and gyroscopes are widely used in the automotive industry to obtain information about the vehicle’s velocity, position and heading by measuring forces and rotations (Elkaim, Lie, & Gebre-Egziabher, 2015; Salychev, 2017). Often found as a set of three accelerometers and three gyroscopes, they produce a six degree of freedom sensor system used in the inertial measurement unit (IMU), the output of which is converted to navigation parameters by the inertial navigation system (INS) (Elkaim et al., 2015). INSs are self-contained non-jammable systems, but suffer errors that have an exponential growth over time and GPS-measurements are used to correct this issue (Spangenberg, Calmettes, & Tourneret, 2007). In the absence of GPS or other external sources of positioning, the vehicle relies on so-called dead-reckoning navigation. Dead-reckoning uses the initial position and calculates the following positions with the use of the IMU, the errors of which can be counteracted with the use of additional sensors such as odometers which alleviates drift, and magnetometers that provide heading and inclination data (Barbour, 2004). Another way to improve localisation performance is by using map-matching techniques (Spangenberg et al., 2007).
Formation tracking for a group of differential-drive mobile robots using an attitude observer
Published in International Journal of Control, 2021
J. González-Sierra, E. Aranda-Bricaire, H. Rodríguez-Cortés, J. Santiaguillo-Salinas
Real-time implementation of control algorithms for differential-drive robots brings up other problems. For instance, the estimation of the position and attitude of a mobile robot is not a simple task (Jakubiak, Lefeber, Tchon, & Nijmeijer, 2002; Noijen, Lambrechts, & Nijmeijer, 2005). Jakubiak et al. (2002) presents two reduced-order observers, one in the case of uncertain position error and the other in the case of unmeasurable orientation. Noijen et al. (2005) uses an observer to estimate the orientation error based on available trajectory information and measurement of the position. Different absolute and relative position estimation approaches have been proposed over the past decade. Amongst the absolute positioning methods, video cameras and ultrasonic emitter-receiver arrangements are frequently encountered. Concerning relative positioning methods, dead reckoning is widely used because of its simplicity. It provides an attitude estimation, but it is unsuitable for long distances due to errors associated with noise and slipping conditions (Borenstein & Feng, 1996). Cedervall and Hu (2007) designs a nonlinear observer based on range sensor readings, assuming that the environment is known. Shi, Yu, and Khoo (2016) study the problem of robust finite-time trajectory tracking of non-holonomic mobile robots with unmeasurable velocities. For this, a composite controller, including observed-based partial state feedback control and disturbance feed-forward compensation, is designed. In Asif, Memon, and Khan (2016), a globally bounded state feedback stabilising controller is proposed for trajectory tracking of a wheeled mobile robot, using a high gain observer for velocity estimation. In the same context, in Liang et al. (2016) and Liang, Wang, Liu, Chen, and Liu (2018), the formation control problem for mobile robots is addressed, where, in the former one, an observer is designed to estimate the local position of the follower robots, using the visual information from a perspective camera, while in the latter one, adaptive observers are designed to estimate the linear velocity of the leader, using image information from an on-board camera.