Imitation learning social navigation method based on feature map fused with pedestrian information

Abstract

The invention discloses an imitation learning social navigation method based on a feature map fused with pedestrian information. According to the imitation learning social navigation method, a robot is guided to imitate the movement habits of experts by introducing an imitation learning method, the navigation method conforming to the social specifications is planned, the planning efficiency is improved, the problem that the robot is locked is solved, and the robot is helped to be better integrated into a man-machine co-integration environment. The method comprises the steps of: acquiring a time sequence motion state of a pedestrian through pedestrian detection and tracking in a sequence RGB image and three-dimensional point cloud alignment; then, in combination with two-dimensional laser data and a social force model, acquiring a local feature map marked with pedestrian dynamic information; and finally, establishing a deep network with the local feature map, a current speed of the robot and a relative position of a target as input and a robot control instruction as output, training with expert teaching data as supervision, and acquiring a navigation strategy conforming to social specifications.

Description

technical field [0001] The invention belongs to the field of mobile robot navigation, and in particular relates to an imitation learning social navigation algorithm based on a feature map fused with dynamic information of pedestrians. Background technique [0002] The positioning of a service robot determines a major feature of its working environment, which is a mixture of humans and machines. From conventional static scenes to human-machine fusion scenes with complex dynamic characteristics, the substantial expansion of the scope of activities puts forward higher requirements for the behavioral norms of robots, which are in line with social norms. On the one hand, service robots must be able to perceive the state of humans in a timely manner through harmonious human-computer interaction, understand human needs, find the best solution, and assist humans in their work with high quality and efficiency. On the other hand, service robots must also be able to While ensuring the...

AI Technical Summary

Problems solved by technology

However, the human-machine sharing environment has high dynamic characteristics, and the complex pedestrian movement destroys the setting conditions of the traditional navigation method. It will be difficult to plan a smooth path in the dense environment if the traditional navigation system is continued, which will affect the comfort of the surrounding pedestrians. degree, and even cause a collision

Images