Predictive navigation of an autonomous vehicle with nonholonomic and minimum turning radius constraints

Augie Widyotriatmo1,홍봉희,홍금식 대한기계학회 2009 JOURNAL OF MECHANICAL SCIENCE AND TECHNOLOGY Vol.23 No.2

A key feature of an autonomous vehicle is the ability to re-plan its motion from a starting configuration (position and orientation) to a goal configuration while avoiding obstacles. Moreover, it should react robustly to uncertainties throughout its maneuvers. We present a predictive approach for autonomous navigation that incorporates the shortest path, obstacle avoidance, and uncertainties in sensors and actuators. A car-like robot is considered as the autonomous vehicle with nonholonomic and minimum turning radius constraints. The results (arcs and line segments) from a shortest-path planner are used as a reference to find action sequence candidates. The vehicle’s states and their corresponding probability distributions are predicted to determine a future reward value for each action sequence candidate. Finally, an optimal action policy is calculated by maximizing an objective function. Through simulations, the proposed method demonstrates the capability of avoiding obstacles as well as of approaching a goal. The regenerated path will incorporate uncertainty information.

Robust stabilization of a wheeled vehicle: Hybrid feedback control design and experimental validation

Widyotriatmo, Augie,Hong, Keum-Shik,Prayudhi, Lafin H. The Korean Society of Mechanical Engineers 2010 JOURNAL OF MECHANICAL SCIENCE AND TECHNOLOGY Vol.24 No.2

In this paper, the problem of robust stabilization of a wheeled vehicle is addressed. The configuration (position and orientation) set of the vehicle is divided into two parts: global and local configuration sets. The novelty of this paper is the design of a hybrid feedback controller that assigns different objectives in the vehicle's global and local behaviors. Two Lyapunov functions for individual objectives are introduced that allow a hybrid feedback control law to pursue different objectives. In the global sense, it is important to reach the target point as quickly as possible, but once the vehicle reaches is near the goal, a precise maneuvering by rejecting disturbances including tire slippage and measurement noise becomes important. The asymptotical stability and robustness of the closed loop system are assured. The derived control law is validated by simulations and experiments using an autonomous forklift.

Navigation Function-Based Control of Multiple Wheeled Vehicles

Widyotriatmo, A,Keum-Shik Hong IEEE 2011 IEEE transactions on industrial electronics Vol.58 No.5

<P>In this paper, a collision-free navigation method for a group of autonomous wheeled vehicles is investigated. The position and orientation information of individual vehicles is transformed to navigation variables, which are the distance left to the goal position, the angle made by the orientation of the vehicle at the goal position and the vehicle-to-target (v-to-t) vector, and the angle made by the heading direction of the vehicle and the v-to-t vector. As a Lyapunov function for deriving a smooth control law that drives all the vehicles from an initial configuration to a goal configuration, a new navigation function that incorporates the squared norm of the navigation variables, the boundaries of collision-free areas, and the angles made by the vehicle heading direction and the vehicle-to-obstacle (v-to-o) vectors is proposed. The asymptotic stability of the closed-loop system is proved. The effectiveness of the developed algorithm is illustrated through three simulations (three vehicles in a free environment, three vehicles in the presence of a static obstacle, and eight vehicles operating along a corridor) and two experiments (static and moving obstacles avoidance). The proposed algorithm has been implemented on a real forklift and the navigation of the forklift from an arbitrary initial configuration to a goal configuration while avoiding collisions has been demonstrated.</P>

Orienting Head-Truck in the Design of Truck-Trailer Path Following Control

Augie Widyotriatmo 제어로봇시스템학회 2018 제어로봇시스템학회 국제학술대회 논문집 Vol.2018 No.10

A method for determining the control algorithm for a truck-trailer system following a certain path is proposed. The control design for path following control of a truck-trailer system is divided into three steps, which are orienting the head-truck, backstepping the steering angle, and position control of steering angle. In the orienting head-truck, the headtruck serves as the steering direction of a truck-trailer system. By designing the orienting head-truck algorithm that, the truck-trailer body is driven to be aligned with the tangent of path-curve. From the orienting head-truck algorithm, a backstepping method is utilized to implement the steering angle control. Then, the position control to set the steering angle to the desired one which is calculated by the steering control. By using this method, the stability of the origin of the distance and orientation error is proven by the Lyapunov method. Simulation results of truck trailer following a rectangular path is performed, showing the effectiveness of the proposed method.

Configuration Control of a Forklift Vehicle using Vision System with Limited Angle of View

Augie Widyotriatmo,Gi-Yong Hong,Keum-Shik Hong 제어로봇시스템학회 2010 제어로봇시스템학회 국제학술대회 논문집 Vol.2010 No.10

This paper presents a configuration control of a forklift vehicle using a camera-based vision system with limited angle-of-view (AOV). The configuration (i.e., position and orientation) of the vehicle is transformed to navigation variables, which are the distance left to the goal position, the angle made by the x-axis of the target coordinate and the vehicle-to-target (v-to-t) vector, and the angle made by the x-axis of the body coordinate of the vehicle and the v-to-t vector. The kinematic equations of the forklift are derived in the navigation variables and the constraint of the motion due to the limited AOV of the camera is formulated. The control law is derived using quadratic and barrier functions as a Lyapunov function candidate. The asymptotical stability of the origin of the navigation variables is assured using the Lyapunov method. From the property of the barrier function, it can be guaranteed that the motion of the vehicle does not violate the constraint due to the limited AOV of the camera. The effectiveness of the proposed method is verified through numerical simulations.

Wall Following Control Algorithm for a Car-Like Wheeled-Mobile Robot with Differential-Wheels Drive

Lafin Hari Prayudhi,Augie Widyotriatmo,Keum-Shik Hong 제어로봇시스템학회 2015 제어로봇시스템학회 국제학술대회 논문집 Vol.2015 No.10

A configuration wheeled mobile robot which is classified as car-like vehicle with a differential wheels drive is developed. The mobile robot is driven based upon differential wheels drive and rotating steer mechanism. The system configuration of mobile robot platform is designed including the necessary sensors and controller. The kinematic model of the mobile robot is derived and the control configuration is proposed. A wall following control algorithm for the robot is developed using the Lyapunov method. The state variables of the wall following model are the distance error from the position of mobile robot to a desired distance from the wall and the orientation error between the orientation of the mobile robot and the wall. The closed-loop system shows asymptotic stabilization of the origin of variables, which are distance and orientation of the mobile robot with respect to the wall. The control algorithm steers the mobile robot to move along the wall with a predetermined distance and zero orientation to the wall. The effectiveness of designed control algorithm is shown using simulation results.

A Motion Planning Algorithm for a Nonholonomic Vehicle Using Vector Potential Functions in Triangular Regions

Anugrah Kusumo Pamosoaji,Augie Widyotriatmo,Keum-Shik Hong 제어로봇시스템학회 2010 제어로봇시스템학회 국제학술대회 논문집 Vol.2010 No.10

A motion planning algorithm for a nonholonomic vehicle in triangular regions is investigated. The regions are the result of splitting a larger and complex workspace, and are classified into three classes, that are, empty regions, obstacle-inside regions, and goal regions. The vehicle has to achieve a goal configuration from any initial configuration in the workspace. A set of procedures to generate velocity vector fields by utilizing vector potential functions is proposed. The vector fields are categorized as those generated by the edges of regions, obstacles, and goal points. To deal with some constraints, i.e., maximum velocities, a set of parameter-scaling rules is provided. A state-feedback controller for a unicycle vehicle is used to show that the generated motion plan can be tracked by the vehicle. Simulation results showing the motion planning from different initial configuration are presented.

Optimization of a Grid-Tied Microgrid Configuration using Dual Storage Systems

Petrus Yuri Nugraha,Augie Widyotriatmo,Edi Leksono 제어로봇시스템학회 2015 제어로봇시스템학회 국제학술대회 논문집 Vol.2015 No.10

This paper presents optimization in capacity sizing and operational schedule for a grid-tied microgrid using dual storage systems. Photovoltaic power system is used as a renewable energy source. Pumped-storage hydroelectricity and battery energy storage system are utilized as energy storage systems. The physical models of the photovoltaic system, pumped-storage hydroelectricity, and battery energy storage system are developed. The objective function is formulated to minimize capital and operational costs of grid-tied microgrid system. The constraints for the optimization are formulated based on the system model, operational limitations, and performance requirements. Capacity and operational schedule for the microgrid system is optimized using solar insolation data and load demand data. Performances required for microgrid system are high renewable energy penetration with low curtailed renewable energy. Mixed integer linear programming (MILP) is used to solve the optimization problem. With the proposed method, renewable energy penetration ratio achieves 50%. The curtailed renewable energy ratio of 10.4% is obtained.

Development and Control Strategies of an Unmanned Autonomous Forklift

홍금식(Keum-Shik Hong),김민환(Min-Hwan Kim),이석(Suk Lee),김성신(Seong-Sin Kim),홍봉희(Bong Hee Hong),위디오트리앗모 아우기(A. Widyotriatmo),브이티탄 규엔(Quyen T. T. Bui),이정훈(Jung Hun Lee),탐바 투아 아구스티누스(Tua A. Tamba) 대한기계학회 2008 대한기계학회 춘추학술대회 Vol.2008 No.5

This paper presents the control strategies for the development of an unmanned autonomous forklift. A control architecture using several sensors (vision, ultrasonic sensor, laser range finder, etc.) and a PC controller is presented. An ultrasonic satellite and gyro are used for the localization, and a vision system is utilized for obstacle detection and the measurement of pallet position. The CAN communication network is utilized to provide the drive-by-wire technology for the unmanned autonomous forklift.