Applying the Kalman filter algorithm on autonomous vehicles: Framework and validation for sensor attack detection and isolation

An enhanced obstacle avoidance algorithm for a network-based autonomous mobile robot is proposed in this paper. Firstly, the readings of the environmental sensors at a moment are compensated to the prospecting readings of the sensors considering network delay measured and the kinematic model of the robot. The compensated readings of the sensors are used for building the polar histogram of the VFH algorithm. Secondly, a sensory fusion using the Kalman filter is proposed for the localization of the robot to compensate both the delay of the readings of an odometry sensor and the delay of the readings of the environmental sensors. The performance enhancements of the proposed obstacle avoidance algorithm from the viewpoint of efficient path generation and accurate goal positioning are also shown in this paper through some simulation tests.

This paper presents the implementation of an autonomous electric vehicle (EV) project in the National Taiwan University of Science and Technology (NTUST) campus in Taiwan. The aim of this work was to integrate two important practices of realizing an autonomous vehicle in a campus environment, including vehicle positioning and path tracking. Such a project is helpful to the students to learn and practice key technologies of autonomous vehicles conveniently. Therefore, a laboratory-made EV was equipped with real-time kinematic GPS (RTK-GPS) to provide centimeter position accuracy. Furthermore, the model predictive control (MPC) was proposed to perform the path tracking capability. Nevertheless, the RTK-GPS exhibited some robust positioning concerns in practical application, such as a low update rate, signal obstruction, signal drift, and network instability. To solve this problem, a multisensory fusion approach using an unscented Kalman filter (UKF) was utilized to improve the vehicle positioning performance by further considering an inertial measurement unit (IMU) and wheel odometry. On the other hand, the model predictive control (MPC) is usually used to control autonomous EVs. However, the determination of MPC parameters is a challenging task. Hence, reinforcement learning (RL) was utilized to generalize the pre-trained datum value for the determination of MPC parameters in practice. To evaluate the performance of the RL-based MPC, software simulations using MATLAB and a laboratory-made, full-scale electric vehicle were arranged for experiments and validation. In a 199.27 m campus loop path, the estimated travel distance error was 0.82% in terms of UKF. The MPC parameters generated by RL also achieved a better tracking performance with 0.227 m RMSE in path tracking experiments, and they also achieved a better tracking performance when compared to that of human-tuned MPC parameters.

An autonomous vehicle is designed to be driven automatically without a crew controlling it. The autonomous vehicle can ease the people in driving. Thus, a control system is important to ensure the autonomous electric vehicle run properly. Various control systems can be implemented on the autonomous vehicle. However, the previous studies have shown that the inaccuracy and inconsistency may still occur. Thus, this study presents the combination of compass and fuzzy logic to control the position of the autonomous vehicle. The Sugeno fuzzy logic method is used to control the position of the movement of autonomous vehicle by considering several inputs, such as the direction to the destination, the position of the destination, and the distance to the intended target. Meanwhile, the rotational speed of the motor becomes the output. In this study, the autonomous vehicle was run in Universitas Sriwijaya located in two locations, namely at Palembang Campus, and Indralaya Campus. The results show that the autonomous vehicle was able to reach the destination position with the motor speed output on the autonomous electric vehicle which was read according to the fuzzy. This study shows that fuzzy rule for 5 memberships function has the lowest heading error of 6.8 degrees compared to 3 memberships function. Nevertheless, 3 memberships function has the shortest time to reach the destination. The results of this study indicate that the control system using the fuzzy logic method works well.

In view of the problems of route selection and driving stability of intelligent vehicle in the process of turning, a state model of intelligent vehicle turning system control based on Kalman filter is proposed. Taking intelligent vehicle as the research object, according to the route of intelligent vehicle passing through a certain bend at uniform speed, the corresponding state transition model of intelligent vehicle’s speed and position is established, and the state model of intelligent vehicle’s bend system is analyzed, and the system state equation of intelligent vehicle in the process of bend movement based on Kalman filter is obtained. According to the state equation of the system, it can effectively monitor the speed, position and trajectory of the intelligent vehicle on the bend. Combining with Kalman filter algorithm, it can effectively reduce the noise of the control process and observation noise. It can be used to achieve more stable and accurate control of the intelligent vehicle in the bend movement.

This paper describes the mathematical model and controller design for Manta-type Unmanned Underwater Test Vehicle (MUUTV) with 6 DOF nonlinear dynamic equations. The mathematical model contains hydrodynamic forces and moments expressed in terms of a set of hydrodynamic coefficients which were obtained through the PMM (Planar Motion Mechanism) test. Based on the 6 DOF dynamic equations, numerical simulations have been performed to analyze the dynamic performances of the MUUTV. In addition, using the mathematical model PID and sliding mode controller are constructed for the diving and steering maneuver. Simulation results show that the control performances of the MUUTV and compared with these of NPS (Naval Postgraduate School) AUV II.Key Words : Manta-type Unmanned Underwater Test Vehicle, Mathematical Model, Controller Design, Depth Control, Heading Control, PID Controller, Sliding Mode Controller 본 논문은 수중운동체특화연구센터(UVRC, SM-23)의 연구비 지원으로 수행되었음. * 교신저자 : 김준영(jkim@jejunu.ac.kr)접수일 09년 10월 15일 수정일 09년 12월 21일 게재확정일 10년 01월 20일

Underwater positioning and navigation for autonomous vehicles is challenging because of the GPS-denied environments. We present a novel positioning and navigation method for underwater vehicles using the Earth's magnetic field properties including field intensity, declination, and inclination. In other words, we assume that the underwater vehicle is equipped with magnetometers that can sense the magnetic field properties at the sampling location. Based on the sensor readings, we estimate the geographical location of the vehicle. Specifically, we develop a feed-forward neural network model to estimate the geographical location of the vehicle by mapping magnetometer sensor readings to a unique location on the Earth's surface. To train the neural network, we generate data using the International Geomagnetic Reference Field (IGRF) model.

<div class="section abstract"><div class="htmlview paragraph">A two-passenger all-wheel-drive urban electric vehicle (AUTO21EV) with four in-wheel motors and an active steering system has been designed and developed at the University of Waterloo. In order to evaluate the handling and performance of such a vehicle in the design stage and analyze the effectiveness of different chassis control systems before implementing them in the real vehicle, the simulation of a large number of different open-loop and closed-loop test maneuvers is necessary. Thus, in the simulation environment, not only is a mathematical vehicle model needed for every test maneuver, but a driver model must also be designed to simulate the closed-loop test maneuvers. The role of the driver model is to calculate the control inputs required to successfully follow a predefined path. Such a driver model can be implemented as an inverse dynamics problem or by a representation of a driver that can look ahead, preview the path, and change the steering wheel angle and acceleration or brake pedal positions accordingly. In this regard, a path-following driver model is developed in this work with an advanced path previewing technique. In addition, a gain scheduling speed control driver model is developed for the AUTO21EV, which adjusts the drive torques of the wheels to minimize the deviation between the desired and actual vehicle speeds.</div></div>

A nonholonomic control method is considered for stabilizing all attitudes and positions (x, y, or z) of an underactuated X4 autonomous underwater vehicle (AUV) with four thrusters and six degrees of freedom (DOF), in which the positions are stabilized according to the Lyapunov stability theory. A dynamic model is first derived, and then a sequential nonlinear control strategy is implemented for the X4-AUV which is composed of translational and rotational subsystems. A controller for the translational subsystem stabilizes one position out of the x-, y-, and z-coordinates, whereas controllers for the rotational subsystems generate the desired roll, pitch, and yaw angles. Thus, the rotational controllers stabilize all the attitudes of the X4-AUV at the desired (x-, y-, or z-) position of the vehicle. Some numerical simulations are conducted to demonstrate the effectiveness of the proposed controllers.

This study presents a field-programmable gate array (FPGA)-based mechatronic design and real-time fuzzy control method with computational intelligence optimization for omni-Mecanum-wheeled autonomous vehicles. With the advantages of cuckoo search (CS), an evolutionary CS-based fuzzy system is proposed, called CS-fuzzy. The CS’s computational intelligence was employed to optimize the structure of fuzzy systems. The proposed CS-fuzzy computing scheme was then applied to design an optimal real-time control method for omni-Mecanum-wheeled autonomous vehicles with four wheels. Both vehicle model and CS-fuzzy optimization are considered to achieve intelligent tracking control of Mecanum mobile vehicles. The control parameters of the Mecanum fuzzy controller are online-adjusted to provide real-time capability. This methodology outperforms the traditional offline-tuned controllers without computational intelligences in terms of real-time control, performance, intelligent control and evolutionary optimization. The mechatronic design of the experimental CS-fuzzy based autonomous mobile vehicle was developed using FPGA realization. Some experimental results and comparative analysis are discussed to examine the effectiveness, performance, and merit of the proposed methods against other existing approaches.

Localization methods for autonomous construction vehicles include real‐time kinematic positioning through a global navigation satellite system (GNSS) and a scan matching with a light detection and ranging (LiDAR) attached to mobility. However, these conventional methods have low estimation accuracy when the vehicle's surroundings have few features and the vehicle is in the no‐GNSS area. For the estimation in such areas, this paper proposes a localization method that can estimate by matching a 3D model of a construction vehicle with a point cloud obtained from 3D LiDARs installed in a work area. To realize the high‐accuracy and high‐speed processing of the localization, we propose remodeling using the predictive motion model (RM) algorithm to modify the 3D model in the registration process. In the experimental results on rough terrain, we confirmed that our method can estimate a vehicle's position and yaw angle with accuracies of 0.121 m and 0.016 rad, respectively. In addition, compared with the case without the RM algorithm, the construction vehicle's position and yaw angle accuracies improved up to 5 and 12 times, respectively.

A vision/inertia integrated positioning method using position and orientation matching which can be adopted on intelligent vehicle such as automated guided vehicle (AGV) and mobile robot is proposed in this work. The method is introduced firstly. Landmarks are placed into the navigation field and camera and inertial measurement unit (IMU) are installed on the vehicle. Vision processor calculates the azimuth and position information from the pictures which include artificial landmarks with the known direction and position. Inertial navigation system (INS) calculates the azimuth and position of vehicle in real time and the calculated pixel position of landmark can be computed from the INS output position. Then the needed mathematical models are established and integrated navigation is implemented by Kalman filter with the observation of azimuth and the calculated pixel position of landmark. Navigation errors and IMU errors are estimated and compensated in real time so that high precision navigation results can be got. Finally, simulation and test are performed, respectively. Both simulation and test results prove that this vision/inertia integrated positioning method using position and orientation matching has feasibility and it can achieve centimeter-level autonomic continuous navigation.

The real-time position of an autonomous underwater vehicle (AUV) is always of great interest. With advances of technologies, underwater wireless sensor networks (UWSNs) become promising tools for AUV tracking. Due to the energy constraint of underwater nodes, energy saving is a key issue that affects all aspects of the design of a tracking scheme. In this paper, we propose a novel energy-efficient tracking scheme for an AUV to locate itself in time by UWSNs. We first design a tracking protocol considering the energy consumptions in both the AUV and sensor nodes (SNs). Particularly, the protocol is designed in two aspects: 1) the passive listening mechanism and duty-cycle strategy for the AUV and 2) the detection-based ranging packet transmission for SNs. Since the tracking protocol will inevitably affect the packet delivery between the AUV and SNs, we analyze the packet delivery success rate (PDSR) to shed light on the impact of system parameters on the tracking performance. To cope with non-linearity of the model and the intermittent observations mainly arisen from the effect of the tracking protocol, we adopt two extended versions of the original intermittent Kalman filter for tracking. They are intermittent extended Kalman filter and intermittent unscented Kalman filter. Simulation results demonstrate the effectiveness of the proposed tracking scheme, and reveal that the PDSR analysis provides a design guidance for parameter selection in system configuration.

In the robot navigation problem, noisy sensor data must be filtered to obtain the best estimate of the robot position. The discrete Kalman filter, commonly used for prediction and detection of signals in communication and control problems, has become a popular method to reduce the effect of uncertainty from the sensor data. However, in the domain of robot navigation, sensor readings are not only uncertain, but can also be relatively infrequent compared to traditional signal processing applications. In addition, a good initial estimate of location, critical for Kalman convergence, is often not available. Hence, there is a need for a filter that is capable of converging with a poor initial estimate and many fewer readings than the Kalman filter. To this end, we propose the use of a Recursive Total Least Squares Filter. This filter is easily updated to incorporate new sensor data, and in our experiments converged faster and to greater accuracy than the Kalman filter.

This paper proposes a new sensor deployment strategy based on quantitative cause-effect graph (QCEG) to handle the heterogeneity among the properties of sensors and faults. A QCEG is developed to model the cause-effect relationship between the system faults and sensor readings. A multi-objective optimization is performed to facilitate the monitoring of single-station multistep manufacturing process (SMMP). A stream of fault information model is built to describe the propagation of fault state in the SMMP. By means of state-space transformation, a detection factor is used to provide the initial sensor deployment. The optimal sensor deployment in an SMMP is achieved by an improved shuffled frog leaping algorithm (ISFLA), which minimizes the fault unobservability, maximizes the system stability, and minimizes the cost for the whole system, under the constraints on detectability, stationarity, and limited resources. Two experimental investigations on an assembly unit and a manufacturing unit are conducted to verify the methodology. Comparative studies demonstrate that the proposed QCEG is able to overcome the shortcomings of directed graph (DG) in handling sensor heterogeneity and multiple objectives. As a goal-oriented swarm-intelligence search strategy, the ISFLA performs better than the popular integer programming in dealing with the multi-objective optimization problem.

The subject of this paper is the fusion of measurement data from IMU (Inertial Measurement Unit), GPS (Global Positioning System) sensors and cameras used in autonomous vehicles to locate an autonomous land vehicle. The experimental part presents the design of a fusion filter, which based on the fusion of data from the IMU, GPS, and camera can determine the position, orientation and speed of the land vehicle. The design of the fusion filter was tested using several simulations in the Matlab environment, the results of which are analyzed at the end of this paper.KeywordsAutonomous vehiclesGPSIMUMVO