H. Hu and M. Brady. A Bayesian approach to real-time obstacle avoidance for a mobile robot. Autonomous Robots, 1(1):69-92, 1994.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....as well. 3. The real time performance of existing hybrid architectures is still not optimal because the capability of the reactive components has not been fully exploited. In extreme cases, the workload of the high level planning module far exceeds that of the low level reactive module (e.g. [7, 18, 28, 33]) The planning module plans the exact motion path and generates the detailed sequence of actions to be executed by the actuators. The reactive module performs a single task, i.e. obstacle avoidance, by making minor modifications to an otherwise good course of action. A good example of ....

H. Hu and M. Brady. A Bayesian approach to real-time obstacle avoidance for a mobile robot. Autonomous Robots, 1:69--92, 1994.

....learning. However, it needs a human who provides reinforcement to the system. In DAC5 the reinforcement signal that induces the retention of a STM sequence in LTM is triggered through its own reflexes. Bayesian approaches have been applied to robots, for instance, in obstacle avoidance tasks [25] and the processing of noisy sensor data in the context of path planning [31] Contrary to DAC5, however, in these cases the priors included in the model are defined a priori by the designers of the system. Although such an approach might be advantageous from an engineering perspective it does ....

H. Hu and M. Brady. A Bayesian approach to real-time obstacle avoidance for a mobile robot. Autonomous Robots, 1:69-92, 1994.

....moving in the real world. Over the past years, a huge variety of techniques for motion planning has been developed. They can roughly be divided into map based approaches such as road map or cell decomposition techniques (see [11] for an extensive overview) and reactive, sensor based approaches [1, 4, 6, 7, 9, 10, 15]. Goal directed path planning techniques compute paths based on a given map of the environment. Thereby they assume that the environment does not change while the robot is moving. However, when robots are designed to operate in dynamic or populated environments, this assumption is no longer ....

H. Hu and M. Brady. A bayesian approach to real-time obstacle avoidance for a mobile robot. In Autonomous Robots, volume 1. Kluwer Academic Publishers, Boston, 1994.

....as well. 3. The real time performance of existing hybrid architectures is still not optimal because the capability of the reactive components has not been fully exploited. In extreme cases, the workload of the high level planning module far exceeds that of the low level reactive module (e.g. [7, 18, 29, 34]) The planning module plans the exact motion path and generates the detailed sequence of actions to be executed by the actuators. The reactive module performs a single task, i.e. obstacle avoidance, by making minor modi cations to an otherwise good course of action. A good example of ....

H. Hu and M. Brady. A bayesian approach to real-time obstacle avoidance for a mobile robot. Autonomous Robots, 1:69-92, 1994.

....shape, and the sensor data consists of tactile information for a point robot. With a probabilistic model, the robot might infer a posterior probability density, p(e) over environments, which is conditioned on sensor observations, initial conditions, or additional knowledge (e.g. 53] 56] [88], 185] Environment predictability uncertainty Suppose again that the space of environments E is known by the robot; however, in addition, the robot knows its current en10 vironment e 2 E. Predictable motion commands might be given to the robot, but with environment predictability uncertainty, ....

H. Hu and M. Brady. A Bayesian approach to real-time obstacle avoidance for a mobile robot. Autonomous Robots, 1(1):69--92, 1994.

....and the sensor data consists of tactile information for a point robot. With a probabilistic model, the robot might infer a posterior probability density, p(e) over environments, which is conditioned on sensor observations, initial conditions, or additional knowledge (e.g. 14] 15] [28], 68] Type EP uncertainty. Suppose again that the space of environments, E , is known by the robot; however, in addition, the robot knows its current environment e 2 E . Predictable motion commands might be given to the robot, but with Type EP uncertainty future environments cannot be ....

H. Hu and M. Brady. A Bayesian approach to real-time obstacle avoidance for a mobile robot. Autonomous Robots, 1(1):69--92, 1994.

....suffers from two limitations, both of which are critical in environments like the museum. 1) Inability to handle invisible obstacles. Virtually all existing methods for collision avoidance are purely sensor based, i.e. they only consult the robot s sensors to determine collision free motion [13,50,71,78,79,137]. If all obstacles can be sensed, such strategies suffice. However, since some of the obstacles in the museum were invisible, a purely sensor based approach would have been likely to fail. 2) Inability to consider dynamics. With few exceptions [50,137] existing approaches model only the ....

H. Hu, M. Brady, A Bayesian approach to real-time obstacle avoidance for a mobile robot, in: Autonomous Robots, Vol. 1, Kluwer Academic, Boston, 1994, pp. 69--92.

....in situations where a massive congestion of the museum forced the robot to take a detour. 1) Inability to handle invisible obstacles. Virtually all existing methods for collision avoidance are purely sensor based, i.e. they only consult the robot s sensors to determine collision free motion [13,50,71,78,79,137]. If all obstacles can be sensed, such strategies suffice. However, since some of the obstacles in the museum were invisible, a purely sensor based approach would have been likely to fail. 2) Inability to consider dynamics. With few exceptions [50,137] existing approaches model only the ....

H. Hu and M. Brady. A Bayesian approach to real-time obstacle avoidance for a mobile robot. In Autonomous Robots, volume 1, pages 69--92. Kluwer Academic Publishers, Boston, 1994.

....of temporal lookahead, which limits their use to rather small lookaheads. Dean et al. 1993] use robot navigation as an example to describe a planning and monitoring algorithm that uses a totally observable Markov model, which assumes that the location of the robot is always known precisely. [Hu and Brady, 1994] use Bayesian techniques to detect unforeseen obstacles in an otherwise completely known environments. type: corridor distance: 2 m with prob. 0.30 3 m with prob. 0.50 4 m with prob. 0.20 topo node topo node type: topo node north: open west: open south: open east: wall Figure 1: Augmented ....

H. Hu and J.M. Brady. A Bayesian approach to real-time obstacle avoidance for a mobile robot. Autonomous Robots, 1(1):69--92, 1994.

....number of ill shaped obstacles prohibited the use of purely sensor based methods for collision avoidance. I. INTRODUCTION In order to operate safely in populated environments, many successful mobile robot systems rely on fast, sensor based collision avoidance modules to control the robot (see e.g. [12, 2, 7, 16, 11, 9]) The predominant paradigm of these approaches is strictly sensor based: Sensor readings are continuously analyzed to determine collision free motion. Unfortunately, the sensor based paradigm has important limitations. If the environment is complex, it might be difficult to equip a robot with a ....

H. Hu and M. Brady. A bayesian approach to real-time obstacle avoidance for a mobile robot. In Autonomous Robots, volume 1. Kluwer Academic Publishers, Boston, 1994.

....not appropriate for fast obstacle avoidance. Their strength is global path planning. Additionally, these methods have proven problematic when the global world model is inaccurate, or simply not available, as is typically the case in most populated indoor environments. Hu Brady, Moravec and others [5, 10], have shown how to update global world models based on sensory input, using probabilistic representations. A second disadvantage of global methods is their slowness due to the inherent complexity of robot motion planning [11] This is particularly problematic if the underlying world model changes ....

Huosheng Hu and Michael Brady. A bayesian approach to real-time obstacle avoidance for a mobile robot. In Autonomous Robots, volume 1, pages 69--92. Kluwer Academic Publishers, Boston, 1994.

....of local minima persists. The dynamic window approach has been integrated with a gross motion planner [13] and was extended to use a map in conjunction with sensory information to generate collision free motion [7] A Bayesian approach to obstacle avoidance was linked with global path planning [8]. However, these approaches require a priori knowledge about the environment for the execution of a motion command. 2.2 Motion Planning There is a large number of robot motion planning algorithms presented in the literature [12] In lowdimensional con guration spaces, like those for mobile robots, ....

Huosheng Hu and Michael Brady. A bayesian approach to real-time obstacle avoidance for a mobile robot. Autonomous Robots, 1:69-92, 1994.

....building with small unforeseen obstacles, while localizing itself in order not to get lost. Temporal belief networks with a limited lookahead are used to derive which action to execute next, based on sensor reports that (probabilistically) differentiate locally distinct places. In similar work, [4] assumes that unforeseen obstacles might block the robot s path in an otherwise completely known environments. A decision theoretic approach is used to interpret sonar sensor reports and to decide whether the current path is blocked. While these approaches are similar to ours in their Bayesian ....

H. Hu and J.M. Brady. A Bayesian approach to real-time obstacle avoidance for a mobile robot. Autonomous Robots, 1995. (submitted).

....in situations where a massive congestion of the museum forced the robot to take a detour. 1. Inability to handle invisible obstacles. Virtually all existing methods for collision avoidance are purely sensor based, i.e. they only consult the robot s sensors to determine collision free motion [10, 42, 62, 68, 69, 118]. If all obstacles can be sensed, such strategies suffice. However, since some of the obstacles in the museum were invisible, a purely sensor based approach would have been likely to fail. 2. Inability to consider dynamics. With few exceptions [42, 118] existing approaches model only the ....

H. Hu and M. Brady. A Bayesian approach to real-time obstacle avoidance for a mobile robot. In Autonomous Robots, volume 1, pages 69--92. Kluwer Academic Publishers, Boston, 1994.

....appropriate for fast obstacle avoidance. Their strength is global path planning. More specifically, these methods have proven problematic when the global world model is inaccurate, or simply not available, as is typically the case in most populated indoor environments. Hu Brady, Moravec and others [5, 9], have shown how to update global world models based on sensory input, using probabilistic representations. A second disadvantage of global methods is their slowness due to the inherent complexity of robot motion planning [10] This is particularly problematic if the underlying world model changes ....

Huosheng Hu and Michael Brady. A Bayesian approach to real-time obstacle avoidance for a mobile robot. In Autonomous Robots, volume 1, pages 69-- 92. Kluwer Academic Publishers, Boston, 1994.

....not appropriate for fast obstacle avoidance. Their strength is global path planning. Additionally, these methods have proven problematic when the global world model is inaccurate, or simply not available, as is typically the case in most populated indoor environments. Hu Brady, Moravec and others [5, 11], have shown how to update global world models based on sensory input, using probabilistic representations. A second disadvantage of global methods is their slowness due to the inherent complexity of robot motion planning [12] This is particularly problematic if the underlying world model changes ....

H. Hu and M. Brady. A bayesian approach to real-time obstacle avoidance for a mobile robot. In Autonomous Robots, volume 1, pages 69--92. Kluwer Academic Publishers, Boston, 1994.

.... a manipulation problem) and the action, to yield e k 1 = f 00 k (x k ; a k ) in which x k = q k e k ] If the current environment is unknown, then there is uncertainty in environment sensing, which is a problem that has been considered in robotics from several different perspectives (e.g. [30, 35, 52, 65, 99]) This can be modeled by defining y k = h k (x k ; s k ) in which x k = q k e k ] In general, sensing and predictability uncertainties can be defined for any state space, including those that include dynamics. Also, a set of parameters could characterize variations in the model, and used to ....

H. Hu and M. Brady. A Bayesian approach to real-time obstacle avoidance for a mobile robot. Autonomous Robots, 1(1):69--92, 1994.

....result in nonfeasible trajectories i.e. trajectories that do not satisfy the constraints on the configuration variables. More recently, researchers have been examining nonholonomic path planning in the presence of obstacles [ Laumond, 1990; Barraquand and Latombe, 1989; Mirtich and Canny, 1992; Hu and Brady, 1995 ] However, while most of these planners provide some excellent results they are quite rigid in the choice of control laws used to steer the robots and often do not exploit the control laws available in control literature, for example [ Murray and Sastry, 1990; Sussmann, 1991; Coron, 1992; de Wit ....

H. Hu and M. Brady. A bayesian approach to real-time obstacle avoidance for a mobile robot. Autonomous Robots, 1(1):69--92, 1995.

No context found.

H. Hu and M. Brady. A Bayesian approach to real-time obstacle avoidance for a mobile robot. Autonomous Robots, 1(1):69-92, 1994.

No context found.

H. Hu and M. Brady. A Bayesian approach to real-time obstacle avoidance for a mobile robot. Autonomous Robots, 1(1):69--92, 1994.

No context found.

H. Hu and M. Brady. A Bayesian approach to real-time obstacle avoidance for a mobile robot. Autonomous Robots, 1(1):69--92, 1995.

No context found.

H. Hu and M. Brady. A Bayesian approach to real-time obstacle avoidance for a mobile robot. Autonomous Robots, 1(1):69--92, 1994.

No context found.

H. Hu and M. Brady, "A Bayesian Approach to Real-time Obstacle Avoidance for Mobile Robots", Autonomous Robots, 1, 1994, pp. 69-92.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC