Sensor-based Navigation Algorithm Research Articles

Sensor-based navigation of a car-like robot based on Bug family algorithms

This article presents a sensor-based navigation algorithm for 3-degree-of-freedom car-like robots with a nonholonomic constraint. Similar to the classical Bug family algorithms, the proposed algorithm enables the car-like robot to navigate in a completely unknown environment by only using range sensor information. The car-like robot uses the local range sensor view to determine the local path so that it moves toward the goal. To guarantee that the car-like robot can approach the goal, the two modes of motion are repeated, termed as motion-to-goal and wall-following. The motion-to-goal behavior lets the robot directly move toward the goal and the wall-following behavior makes the car-like robot circumnavigate the obstacle boundary until it meets the leaving condition. For each behavior, the nonholonomic motion for the car-like robot is planned in terms of the turning radius at each step. The proposed algorithm is implemented with a real robot and the experimental results show the performance of the proposed algorithm.

A Navigation Algorithm for Avoidance of Moving and Stationary Obstacles for Mobile Robot

Conventional sensor-based navigation algorithms for mobile robot such as Tangent Bug algorithm work only for stationary obstacles. When a mobile robot is operated in more general unknown environment, moving obstacles such as human should also be considered. In this paper, a navigation algorithm which works for moving obstacles and stationary ones with unknown environment is proposed, using a new idea that distinguishes moving obstacles from stationary ones with distance information by sensor of mobile robot. The idea is based on the definition of an inclination angle on the wall surface that is called the wall surface angle. The wall surface angle of each step is accumulated while following the boundary. When the difference between the total accumulated angle and the initial one that is the rotation angle of the robot is over 2π radian, currently following obstacle is recognized as a moving obstacle. According to this idea with Tangent Bug algorithm for stationary obstacles, the navigation algorithm for moving convex shaped obstacles is constructed. In this paper the effectiveness of this algorithm for moving obstacles is shown by simulations.

1A1-C04 移動ロボットの移動障害物回避アルゴリズム : 移動障害物と固定障害物の混在する未知環境でのゴールへの収束性(複数ロボットの協調制御)

A sensor-based navigation algorithm for mobile robot such as Tangent Bug algorithm works only for stationary obstacles. In this paper we proposed the navigation algorithm which works for moving obstacles and stationary ones with unknown environment, using a new idea that distinguishes moving obstacles from stationary ones with a distance information of the sensor of mobile robot, while moving along the boundary of the obstacles. The idea is based on the definition of an angle toward the wall surface that is called the direction vector toward the wall surface. The angle of the direction vector of each step is accumulated while following the boundary. When the total accumulated angle is over 2π radian, the following obstacle is recognized a moving obstacle. In this paper the effectiveness of this algorithm for moving obstacles is shown by some simulations.

Dynamic Modeling and Simulation of a Wheeled Mobile Robot for Traversing Uneven Terrain Without Slip

It is known in literature that a wheeled mobile robot (WMR), with fixed length axle, will undergo slip when it negotiates an uneven terrain. However, motion without slip is desired in WMR’s, since slip at the wheel-ground contact may result in significant wastage of energy and may lead to a larger burden on sensor based navigation algorithms. To avoid slip, the use of a variable length axle (VLA) has been proposed in the literature and the kinematics of the vehicle has been solved depicting no-slip motion. However, the dynamic issues have not been addressed adequately and it is not clear if the VLA concept will work when gravity and inertial loads are taken into account. To achieve slip-free motion on uneven terrain, we have proposed a three-wheeled WMR architecture with torus shaped wheels, and the two rear wheels having lateral tilt capability. The direct and inverse kinematics problem of this WMR has been solved earlier and it was shown by simulation that such a WMR can travel on uneven terrain without slip. In this paper, we derive a set of 27 ordinary differential equations (ODE’s) which form the dynamic model of the three-wheeled WMR. The dynamic equations of motion have been derived symbolically using a Lagrangian approach and computer algebra. The holonomic and nonholonomic constraints of constant length and no-slip, respectively, are taken into account in the model. Simulation results clearly show that the three-wheeled WMR can achieve no-slip motion even when dynamic effects are taken into consideration.

移動ロボットのセンサベーストナビゲーションについて

The sensor-based navigation is the problem of how to select a sequence of sensor-based behaviors between start and goal positions of a mobile robot. If the robot does not know its 2-D environment completely or partially, it is obliged to rely on sensor information reflected from closer obstacles in order to avoid them in an on-line manner. In the on-line framework, we should consider how a mobile robot reaches its goal position in an uncertain 2-D environment. In this paper, we survey almost all the previous sensor-based navigation algorithms, classify their convergences of a mobile robot to its goal position into four categories by basic conditions, and then discuss about acceleration of the convergences by additional conditions.

On the Sensor-Based Navigation Under Position and Orientation Errors

A mobile robot usually estimates its position and orientation by a dead reckoning system. However under several kinds of uncertainties, e.g. inner/outer sensor errors and/or a slip, a mobile robot unfortunately misunderstands true position and orientation in a real world as false position and orientation in a computer world. Due to the difference, a mobile robot governed by sensor-based navigation sometimes gets lost in an unknown 2-D environment. On the observation, in this paper, we firstly improve previous metric sensor-based navigation algorithms in order for a mobile robot to deal with such uncertainties. Secondly, we name a class of errors a homeomorpic error, where true and false positions in real and computer worlds are projected each other by the homeomorphism. Finally, we theoretically show that generation of the homeomorpic error is the necessary and sufficient condition for a mobile robot not to enter into any limit cycle in every type of sensor-based navigation algorithm.

Development of Sensor-Based Navigation for Mobile Robots Using Target Direction Sensor

Sensor-based navigation used a target direction sensor for mobile robots among unknown obstacles in work space is discussed. The advantage of target direction information is robustness of measurement error for online navigation, compared to robot location information. Convergence of navigation using target direction information is discussed. An actual sensor system using two CdS sensors to measure target direction is proposed. Using target direction information, we present a new sensor based navigation algorithm in unknown obstacle environment. The navigation algorithm is based on target direction information, unlike sensor-based motion planning algorithms based on mobile robot location information. Using a sensor-based navigation system, we conducted a navigation experiment and simulations in unknown obstacle environment.

Special Issue on Mobile Robot

Mobility, or locomotion, is as important a function for robots as manipulation. A robot can enlarge its work space by locomotion. It can also recognize its environment well with its sensors by moving around and by observing its surroundings from various directions. Much researches has been done on mobile robots and the research appears to be mature. Research activity on robot mobility is still very active; for example, 22% of the sessions at ICRA'98 - the International Conference on Robotics and Automation - and 24% of the sessions at IROS'98 - the International Conference on Intelligent Robots and Systems - dealt with issues directly related to mobile robots. One of the main reasons may be that intelligent mobile robots are thought to be the closest position to autonomous robot applications. This special issue focuses on a variety of mobile robot research from mobile mechanisms, localization, and navigation to remote control through networks. The first paper, entitled ""Control of an Omnidirectional Vehicle with Multiple Modular Steerable Drive Wheels,"" by M. Hashimoto et al., deals with locomotion mechanisms. They propose an omnidirectional mobile mechanism consisting of modular steerable drive wheels. The omnidirectional function of mobile mechanisms will be an important part of the human-friendly robot in the near future to realize flexible movements in indoor environments. The next three papers focus on audiovisual sensing to localize and navigate a robot. The second paper, entitled ""High-Speed Measurement of Normal Wall Direction by Ultrasonic Sensor,"" by A. Ohya et al., proposes a method to measure the normal direction of walls by ultrasonic array sensor. The third paper, entitled ""Self-Position Detection System Using a Visual-Sensor for Mobile Robots,"" is written by T. Tanaka et al. In their method, the position of the robot is decided by measuring marks such as name plates and fire alarm lamps by visual sensor. In the fourth paper, entitled ""Development of Ultra-Wide-Angle Laser Range Sensor and Navigation of a Mobile Robot in a Corridor Environment,"" written by Y Ando et al., a very wide view-angle sensor is realized using 5 laser fan beam projectors and 3 CCD cameras. The next three papers discussing navigation problems. The fifth paper, entitled ""Autonomous Navigation of an Intelligent Vehicle Using 1-Dimensional Optical Flow,"" by M. Yamada and K. Nakazawa, discusses navigation based on visual feedback. In this work, navigation is realized by general and qualitative knowledge of the environment. The sixth paper, entitled ""Development of Sensor-Based Navigation for Mobile Robots Using Target Direction Sensor,"" by M. Yamamoto et al., proposes a new sensor-based navigation algorithm in an unknown obstacle environment. The seventh paper, entitled ""Navigation Based on Vision and DGPS Information for Mobile Robots,"" S. Kotani et al., describes a navigation system for an autonomous mobile robot in an outdoor environment. The unique point of their paper is the utilization of landmarks and a differential global positioning system to determine robot position and orientation. The last paper deals with the relationship between the mobile robot and computer networks. The paper, entitled ""Direct Mobile Robot Teleoperation via Internet,"" by K. Kawabata et al., proposes direct teleoperation of a mobile robot via the Internet. Such network-based robotics will be an important field in robotics application. We sincerely thank all of the contributors to this special issue for their cooperation from the planning stage to the review process. Many thanks also go to the reviewers for their excellent work. We will be most happy if this issue aids readers in understanding recent trends in mobile robot research and furthers interest in this research field.

Navigation Algorithm for Mobile Robots Using Information of Target Direction.

This paper proposes a sensor-based navigation algorithm for mobile robots using information of target direction. The mobile robot moves from an initial position to a desired target position in obstructed environment by the navigation algorithm, where the robot does not know positions and shapes of obstacles, but only knows local information on obstacle's wall and target direction by it's sensor. The algorithm uses target direction instead of robot's position information which is used in other sensor-based navigation algorithm. This navigation method has robustness on the measurement error of target direction, while other navigation algorithms may not be able to work if the robot's position has measurement error. The main point of the algorithm is using the rotational angle of the robot based on topological characteristic of obstacles. Some simulations using the sensorbased navigation algorithm show an effectiveness by comparison of the proposed method and a conventional position based algorithm. Experimental navigation system using the proposed algorithm is developed for confirming the validity of the simulation study.

Navigation of an Autonomous Mobile Robot by Coordination of Behaviors

In this paper we propose a sensor-based navigation algorithm based on the coordination of elementary behaviors, mainly built on fuzzy inference systems. The proposed navigator combines two types of obstacle avoidance behaviors, one for the convex obstacles and one for the concave ones. To avoid the convex obstacles the navigator fuses a "reaching the middle of a collision-free space" behavior and a "goal-seeking" behavior. The fust one is based on a fuzzy inference system and the second one uses a force derived from an attractive potential field. The concave obstacle avoidance behavior results of the mixture of a "wall-following" behavior with a creation of transition subgoals. An automatically on-line tuned fuzzy wall-following system using a neuro-fuzzy structure is designed. The effectiveness of the proposed method is verified by a series of experiments on the miniature mobile robot Khepera®.

On a Sensor-Based Navigation for a Mobile Robot

The sensor-based navigation for a mobile robot is a problem of how to select a sequence of sensor-based behaviors between start and goal positions. If a mobile robot does not know its 2-d environment completely or partially, it is obliged to rely on sensor information reflected from closer obstacles in order to avoid them on-line. In the on-line framework, we should consider how a mobile robot reaches its goal position in an uncertain 2-d world. Therefore we will study some previous sensor-based navigation algorithms for mobile robots. Our motivations are to ascertain the convergence of a mobile robot to its goal position, compare the lengths of sensor-based sequences made in the previous algorithms, and decrease the length of a sequence of sensor-based motions, which is generated between start and goal positions by a sensor-feedback obstacle avoidance. Because the mobile robot itself, its sensors, and its environment usually have several uncertainties, it is notable as to how a mobile robot arrives at or near its goal in overcoming such uncertainties. It is demonstrated that the sensor-based navigation still has an enormous potential as an actual navigation of a mobile robot in a completely or partially unknown environment.