Collision avoidance

About: Collision avoidance is a research topic. Over the lifetime, 8014 publications have been published within this topic receiving 111414 citations.

Integrated control for pHRI: Collision avoidance, detection, reaction and collaboration

Abstract: We present an integrated control framework for safe physical Human-Robot Interaction (pHRI) based on a hierarchy of consistent behaviors. Safe human robot coexistence is achieved with a layered approach for coping with undesired collisions and intended contacts. A collision avoidance algorithm based on depth information of the HRI scene is used in the first place. Since collision avoidance cannot be guaranteed, it is supported by a physical collision detection/reaction method based on a residual signal which needs only joint position measures. On top of this layer, safe human-robot collaboration tasks can be realized. Collaboration phases are activated and ended by human gestures or voice commands. Intentional physical interaction is enabled and exchanged forces are estimated by integrating the residual with an estimation of the contact point obtained from depth sensing. During the collaboration, only the human parts that are designated as collaborative are allowed to touch the robot while, consistently to the lower layers, all other contacts are considered undesired collisions. Preliminary experimental results with a KUKA LWR-IV and a Kinect sensor are presented.

Flying Fast and Low Among Obstacles: Methodology and Experiments

Abstract: Safe autonomous flight is essential for widespread acceptance of aircraft that must fly close to the ground. We have developed a method of collision avoidance that can be used in three dimensions in much the same way as autonomous ground vehicles that navigate over unexplored terrain. Safe navigation is accomplished by a combination of online environmental sensing, path planning and collision avoidance. Here we outline our methodology and report results with an autonomous helicopter that operates at low elevations in uncharted environments, some of which are densely populated with obstacles such as buildings, trees and wires. We have recently completed over 700 successful runs in which the helicopter traveled between coarsely specified waypoints separated by hundreds of meters, at speeds of up to 10 m s—1 at elevations of 5—11 m above ground level. The helicopter safely avoids large objects such as buildings and trees but also wires as thin as 6 mm. We believe this represents the first time an air vehicle has traveled this fast so close to obstacles. The collision avoidance method learns to avoid obstacles by observing the performance of a human operator.

Highway Capacity Benefits from Using Vehicle-to-Vehicle Communication and Sensors for Collision Avoidance

Abstract: Several automobile manufacturers are offering assisted driving systems that use sensors to automatically brake automobiles to avoid collisions. Before extensively deploying these systems, we should determine how they will affect highway capacity. The goal of this paper is to compare the highway capacity when using sensors alone and when using sensors and vehicle-to-vehicle communication. To achieve this goal, the rules for using both technologies to prevent collisions are proposed, and highway capacity is estimated based on these rules. We show that both technologies can increase highway capacity. The increase in capacity is a function of the fraction of the vehicles that use a technology. If all of the vehicles use sensors alone, the increase in highway capacity is about 43%. While if all of the vehicles use both sensors and vehicle-to-vehicle communication, the increase is about 273%.

Application of social force model to pedestrian behavior analysis at signalized crosswalk

Abstract: Limited pedestrian behavior models shed light on the case at signalized crosswalk, where pedestrian behavior is characterized by group or individual evasion with surrounding pedestrians, collision avoidance with conflicting vehicles, and response to signal control and crosswalk boundary. This study fills this gap by developing a microscopic simulation model for pedestrian behavior analysis at signalized intersection. The social force theory has been employed and adjusted for this purpose. The parameters, including measurable and non-measurable ones, are either directly estimated based on observed dataset or indirectly derived by maximum likelihood estimation. Last, the model performance was confirmed in light of individual trajectory comparison between estimation and observation, passing position distribution at several cross-sections, collision avoidance behavior with conflicting vehicles, and lane-formation phenomenon. The simulation results also concluded that the model enables to visually represent pedestrian crossing behavior as in the real world.