Nor Hashim Mohd Arshad

An analytical process of 2D sonar sensor low altitude field mapping for UAV application

Analytical processes is performed for calculating point distance using ultrasonic sensors, with the intention to resolve a low-altitude two-dimensional surface mapping using UAV. An experimental structure model was constructed to allow for dynamic one-dimensional movement of the sensor system in a horizontal plane. Sample experimental surfaces are chosen to be scanned by the sensor system and recreation of the two-dimensional profile was implemented using a 16-bit microcontroller unit. The recreated profiles are shown for comparison between the calculated and actual values, which proved to be nearly similar.

A study of Kalman's filter in embedded controller for real-time quadrocopter roll and pitch measurement

Autonomous quadrocopter recently has shown a significant usages and potentials to be used widely in the commercial market. This caused the aircraft to become a well-liked subject among the researchers and research organizations. With good controllability and stability on its motions will make the usability of the quadrocopter to perform a task reliably. For this to happen, clean signals from accelerometer and gyroscope sensors appreciably play an essential role. However, the sensors signals are routinely encounter problems with noise and drift. Hence, a filter is required to correct this imperfectness. In this paper, a simplified algorithm for Kalmans filter appropriates to be used by a low-cost microcontroller is presented. The research mainly focuses on filtering noisy outputs from the accelerometer so that the output is necessarily clean to be used for real-time quadrocopter pitch and roll angles estimation. Experimental results on assessment of the algorithm by varying the process and measurement noises are provided. The results reveal with correct tuning on the noise parameters; the filter algorithm is satisfying to give stable angles values with fine accuracy.

Development of an energy efficient driving strategy for a fuel cell vehicle over a fixed distance and average velocity

This paper examines the energy efficiency of a hydrogen fuel cell vehicle operating at different power settings. The goal is to develop a driving strategy to maximize energy efficiency for a fixed distance and average velocity. The test vehicle is equipped with a proton exchange membrane (PEM) fuel cell system that provides electric power to a brushless DC motor. This vehicle was designed to compete in the Shell Eco Marathon, an international competition in which the winners are the teams that go the furthest using the least amount of energy. A computer model that simulated the motor and vehicle dynamics was used to predict the output power demanded from the fuel cell. An actual test was then conducted to verify, compare and analyze the performance of the motor for various speed ranges. The result showed how the efficiency varied for different vehicle accelerations. This data can then be used as a basis to operate the vehicle with optimal efficiency.

Real-time vision based sensor implementation on unmanned aerial vehicle for features detection technique of low altitude mapping

Feature detection is an important task for an autonomous vehicle in getting reference or guide in a agriculture field. Recently, a lot of sensors were implemented for feature detection of a field such as sonar sensor, laser range finder, infrared sensor which it is used as height estimation, size area measurement and path tracking that act as reference for desired measurement in creating a virtual map of the agriculture field which can be used as autonomous vehicle navigation process. A vision based sensor is implemented on the unmanned area vehicle for low altitude mapping. Real-time process with minimum computational process and light weight devices is required and utilized as it used UAV as the transportation element. An experimental setup utilized an analog camera as the front end sensor for the desired detection. The experimental surface is scanned by the analog vision sensor for sensor accuracy analysis used for feature detection application. The process technique is implemented via the medium speed 16-bit processor microcontroller which avoids the requirement of high-speed processor to do image processing. The vision sensor will be utilized into an unmanned aerial vehicle used for feature detection scheme for field measurement. The composing image produced is shown for comparison with the actual image as for performance analysis of the vision based sensor. The designed technique gave acceptable error of image accuracy that adequate to use for features detection on unmanned aeria vehicle.

Simultaneous localization and mapping based real-time inter-row tree tracking technique for unmanned aerial vehicle

This paper presents a review of previous work in the field of the unmanned autonomic vehicles in the agricultural field, particularly in the real-time inter-row tracking and localization techniques. A new method in row detection techniques based on simultaneous localization and mapping is proposed. The inter-row tracking technique differs from other available approaches done in terms of detection and navigation means. This technique is purposely for small-scaled unmanned aerial vehicle where substantial devices or equipments is unwelcome due to its enormous weight and size. Vision-based, laser-based and stereo vision-based techniques found to be an accurate row detection technique, however, proven to be impractical for on-board UAV detection and navigation implementation, as bulky computer as well as fast and robust processors is necessitated to perform all the massive computations and algorithmic tasks. In the way to maneuver the autonomic vehicles and robots to the admissible way points, the GPS receiver allows for precise in navigation and localization in the previous implementation. Unfortunately, it is impractical in applications where there are no GPS signals or receptions such as in the canopied environment like in the agriculture fields e.g. palm oil, rubber and papaya. Thus, the proposed new technique which applies an uncomplicated algorithmic sensory measurement at two visible landmarks, which are the trees, and uses them to predict the admissible way points for UAV navigation without carrying the burden cause by the substantial computers, monolithic computations and distortion of the GPS signals.

Dynamic Height Measurement using Sonar Sensor Based on Single Dimension Kalman Filter in Embedded Controller

The aim of this paper is to evaluate real-time accurateness of low-altitude height measurement using a single sonar sensor via Kalman filter algorithm in embedded controller. The function of the one dimension embedded Kalman filter of sonar sensor is to reduce and resolve the error of low altitude height measurement that produced from the sensor for mapping application. Series data of varied height measurement with Kalman filter was gathered and implemented using a 16-bit micro controller unit. A structure model was constructed to allow for dynamic height measurement between ground surface and the sensor. The tabulated data are shown for comparison of accuracy and error analysis between both with Kalman and without Kalman filter. After several scanned surfaces and analysis of the recorded data, the technique shows the usefulness in height measurement application that used surface scan method which the sonar project to the ground such as UAV in order to reduce error of the detected varied height.

Self-tuning fuzzy fractional-order PI controller: Design and application in steam distillation process

Fractional-order PID control is currently an emerging control technology. Its overall performance had been proven to be better than PID in many applications. Currently, researchers are looking towards combining the fractional-order PID with fuzzy to make it become a self-tuning fractional-order PID. This research had successfully developed a self-tuning fuzzy PI that has been evaluated on a steam distillation process. The proposed fuzzy structure had produced better performance compared to PI, fractional-order PI and self-tuning fuzzy PI.

A Real-Time deviation detection and vector measurement technique for straight line quadrocopter navigation using accelerometer

In autonomous quadrocopter navigation, wind disturbances commonly pose a problem for the vehicle to navigate precisely on the planned path. The vehicle easily deviates off from the path and obviously affects the navigation accuracy. Trajectory correction is one possible solution proposed by few researchers to avert the problem. The approaches are proven to increase the efficiency for the vehicle to navigate on the path with tolerable accuracy error. Motivate by the solution, it is the intention for the research to advise a new trajectory correction algorithm known as a Real-Time Adaptive Trajectory Correction (RATC) algorithm. However, in this paper, a primary work for the algorithm definitely the deviation detection and vector measurement technique via an accelerometer is presented. The technique is designed based on trigonometric functions and vector synthesizing principle. A simplify Kalmans filter appropriately for low-cost onboard controller is employed to overcome the accelerometer drift and double integration error. The technique is evaluated in terms of measurement accuracy where an X-Y platform is utilized as the navigation path and control environment. The deviation vector is logged wirelessly in the Graphical User Interface (GUI) window which also programmed to exhibit visual deviation occurred on the quadrocopter. Three sets of experiment results with different wind disturbance acceleration are discussed. The results signified the technique has fine accuracy for short-duration measurement, especially navigation in a small-bounded area.

A real-time angle deviation detection and measurement technique for straight line quadrocopter navigation using accelorometer

Deviation from a planned path due to dynamic wind disturbances clearly will defect the quadrocopter navigation accuracy thus result to poor execution of a task. Few researchers proposed correction algorithms to be integrated in the autonomous quadrocopter system for the problem arises. However, the algorithms are not applicable to enable quadrocopter to navigate precisely on a planned trajectory around the small area with presences of dynamic disturbances such as wind. The research proposed a preliminary work for new real-time adaptive trajectory correction algorithm for autonomous quadrocopter, specifically the algorithm for angle deviation detection and measurement technique via an accelerometer. A ±3G analog accelerometer is utilized and being interfaced with a low-cost microcontroller for digitization process. For minimizing the fluctuation of accelerometer discrete signals, a simplified Kalmans filter algorithm appropriated for the controller is employed. The filtered signals are then used to sense the quadrocopter deviation from the planned path and compute the angle. A custom-made X-Y platform is used as the straight navigation path and represented as dynamic control environment. A GUI window is built in order to exhibit the visual deviation experienced by quadrocopter and recorded wirelessly the deviation angle from the controller so that evaluation on the accuracy performance can be analyzed. Three experiments with a different range of wind disturbance acceleration are executed where the results reveal that the technique is applicable to be used for short-duration detection and in a small-bounded area with fine accuracy.

Tree diameter measurement using single infrared sensor for non-stationary vehicle context in agriculture field

One of the important aspects in building an autonomous vehicle with the cognitive capability in agriculture field is the potentiality of it to recognize the trees or significant landmarks and map it for future navigation process. Therefore, it is crucial for a machine to detect the trees and keep tracking between them in order to perform the plantation activities autonomously. One of the challenges faced by this implementation is the need of a real-time tree detection technique which will not burden the processor as well as lightweight hardware that suits the needs of an autonomous aerial vehicle. Due to that, this paper presents a new and relatively simple tree diameter measurement technique using a high-performance and non-intrusive infrared sensor. The experiments are performed with a various coloured pole at a static velocity implementation to develop a promising technique that is useful and promising in this context.