DMCA
The Curvature-Velocity Method for Local Obstacle Avoidance (1996)
Venue: | In Proc. of the IEEE International Conference on Robotics and Automation |
Citations: | 155 - 7 self |
Citations
2504 |
Robot Motion Planning
- Latombe
- 1991
(Show Context)
Citation Context ...nments, this neglect of vehicle dynamics can be critical. While methods that take vehicle dynamics and nonholonomic constraints into account have been studied in the context of off-line path planning =-=[4, 9]-=-, such methods are generally too computationally expensive for fast local obstacle avoidance. Recently, however, several local obstacle avoidance methods have been reported that do incorporate vehicle... |
1345 | Real-time obstacle avoidance for manipulators and mobile robots
- Khatib
- 1986
(Show Context)
Citation Context ...Work Several well-known local obstacle avoidance methods work by computing a direction for the robot to head in, but do not take vehicle dynamics into account. For example, Potential Field approaches =-=[1, 8]-=- use vector sums of repulsive and attractive features to compute a desired robot heading. Speed control is sometimes handled by choosing velocity proportional to the magnitude of the potential vector.... |
484 | The vector field histogram-fast obstacle avoidance for mobile robots. Robotics and Automation
- Borenstein, Koren
- 1991
(Show Context)
Citation Context ...ttractive features to compute a desired robot heading. Speed control is sometimes handled by choosing velocity proportional to the magnitude of the potential vector. The Vector Field Histogram method =-=[2]-=- improves on this approach by computing a one-dimensional polar histogram, which is then processed to detect open areas for the robot to travel through. Robot velocity, chosen after the direction has ... |
433 | The dynamic window approach to collision avoidance,”
- Fox, Burgard, et al.
- 1997
(Show Context)
Citation Context ...ds of travel. The original method used a two-step approach to pick curvatures and velocities; subsequently, they have adopted our one-step method for simultaneously choosing curvatures and velocities =-=[6]-=-. A similar approach has been developed for outdoor navigation [7]. Here, full vehicle dynamics are considered, so the path is not necessarily a circular arc, a traversability measure is calculated fo... |
335 |
Motor schema-based mobile robot navigation
- Arkin
- 1989
(Show Context)
Citation Context ...Work Several well-known local obstacle avoidance methods work by computing a direction for the robot to head in, but do not take vehicle dynamics into account. For example, Potential Field approaches =-=[1, 8]-=- use vector sums of repulsive and attractive features to compute a desired robot heading. Speed control is sometimes handled by choosing velocity proportional to the magnitude of the potential vector.... |
119 | The mobile robot Rhino.
- Buhmann, Burgard, et al.
- 1995
(Show Context)
Citation Context ... the point in velocity space that maximizes the objective function. A similar method for high-speed indoor navigation that operates in velocity space was developed somewhat earlier, but independently =-=[3]-=-. This method looks at a discrete set of arcs, constrained by the vehicle dynamics, and chooses one that most closely heads in the goal direction, while ensuring that the robot does not hit an obstacl... |
108 |
Planning smooth paths for mobile robots
- Jacobs, Canny
- 1989
(Show Context)
Citation Context ...nments, this neglect of vehicle dynamics can be critical. While methods that take vehicle dynamics and nonholonomic constraints into account have been studied in the context of off-line path planning =-=[4, 9]-=-, such methods are generally too computationally expensive for fast local obstacle avoidance. Recently, however, several local obstacle avoidance methods have been reported that do incorporate vehicle... |
88 | Probabilistic navigation in partially observable environments
- Simmons, Koenig
- 1995
(Show Context)
Citation Context ...then the weight for the goal-heading term (a 3 ) is set very high to "strongly encourage" the robot to head towards the goal. These behaviors, in turn, form the basis for map-based navigatio=-=n schemes [10, 11]-=-. The resulting system exhibits Figure 9: Traces of Xavier's Runs CVM: 30 cm/sec CVM: 60 cm/sec Pot Field: 60 cm/sec CVM: 45 cm/secsreliable, and speedy, navigation in peopled office/corridor environm... |
80 | An Intelligent Predictive Control Approach to the High-Speed, Cross Country Autonomous Navigation Problem”,
- Kelly
- 1995
(Show Context)
Citation Context ... curvatures and velocities; subsequently, they have adopted our one-step method for simultaneously choosing curvatures and velocities [6]. A similar approach has been developed for outdoor navigation =-=[7]-=-. Here, full vehicle dynamics are considered, so the path is not necessarily a circular arc, a traversability measure is calculated for each path, and the one with the best value is chosen. Both these... |
49 |
Robust Obstacle Avoidance in Unknown and Cramped Environments",
- Feiten, Bauer, et al.
- 1994
(Show Context)
Citation Context ...ly, however, several local obstacle avoidance methods have been reported that do incorporate vehicle dynamics, choosing steering commands rather than travel direction. The Steering Angle Field method =-=[5]-=-, like our own, uses the curvatures tangent to obstacles to constrain a continuous space (in their case, the one-dimensional space of steering angles). The curvatures and associated arc distances are ... |
24 |
Becoming increasingly reliable
- Simmons
- 1994
(Show Context)
Citation Context ...then the weight for the goal-heading term (a 3 ) is set very high to "strongly encourage" the robot to head towards the goal. These behaviors, in turn, form the basis for map-based navigatio=-=n schemes [10, 11]-=-. The resulting system exhibits Figure 9: Traces of Xavier's Runs CVM: 30 cm/sec CVM: 60 cm/sec Pot Field: 60 cm/sec CVM: 45 cm/secsreliable, and speedy, navigation in peopled office/corridor environm... |