An Unmanned Vehicle Obstacle Avoidance Method Based on the Combination of Gradient Descent Algorithm and Vo Method

Abstract

The invention belongs to the technical field of unmanned surface vehicles and discloses an unmanned surface vehicle obstacle avoidance method based on a gradient descent algorithm and a VO (velocity obstacle) method. According to the method, whether a collision risk exists between an unmanned surface vehicle and an obstacle is judged according to the current position and motion information of theunmanned surface vehicle and the position and speed information of the obstacle; if no collision risks exist between the unmanned surface vehicle and the obstacle, the speed and rudder angle of the current navigation output of the unmanned surface vehicle are adopted as the speed and rudder angle of obstacle avoidance output; and if a collision risk exists between the unmanned surface vehicle andthe obstacle, the speed and rudder angle of the current navigation output of the unmanned surface vehicle, adopted as initial values, are inputted into a gradient descent algorithm program, the speedand rudder angle of the obstacle avoidance output of the unmanned surface vehicle are obtained through the gradient descent algorithm program, so that the unmanned surface vehicle can avoid the obstacle. Since the gradient descent algorithm is adopted to find a local optimal solution, and therefore, the unmanned surface vehicle can avoid the obstacle with a smoother trajectory, and the problem ofthe unsteady navigation of the unmanned surface vehicle which is caused by the abrupt change of two optimal solutions obtained before and after a VO region change during the optimal solution finding of a VO method in the prior art can be solved.

Description

technical field [0001] The invention belongs to the technical field of unmanned boats, and in particular relates to an unmanned boat obstacle avoidance method based on a gradient descent algorithm combined with a VO method. Background technique [0002] The surface unmanned boat, referred to as unmanned boat, is a light intelligent surface vehicle with the characteristics of small size, low cost, fast speed and strong maneuverability. With the advancement of control technology, sensor technology, and wireless communication technology, surface unmanned vehicles have achieved great development. By carrying different equipment, the unmanned boat can be used in different fields. For example, when equipped with sonar equipment such as single beam, multi-beam, and shallow bottom profiler, it can be used for seabed mapping, mine detection and anti-submarine, etc.; when equipped with water quality sampling Or when testing equipment, it can be used for environmental testing; when ca...

AI Technical Summary

Problems solved by technology

The VO method is based on the VO area to obtain the global optimal solution (the optimal obstacle avoidance speed and rudder angle). Since the VO area obtained by the VO method will change with the relative position of the unmanned vehicle and the obstacle, when If the VO area changes, the global optimal solution obtained based on the VO area will also change. The optimal algorithm will find the global optimal solution, and the VO method corresponds to more than one minimum value, resulting in the optimal solution obtained twice before and after. The solutions correspond to different minimum values, resulting in sudden changes in the optimal solutions obtained twice (that is, the speed and rudder angle of the two obstacle avoidance outputs change in a large range), which leads to the unstable navigation of the unmanned vehicle ( see figure 2 , figure 2 It is the distribution curve of the cost function in the velocity space corresponding to the two adjacent frames of the VO method obstacle avoidance simulation experiment. From this figure, we can find that there are two extreme points in the cost function space, and the optimal value is obtained when the cost function changes. The value will jump at two extreme points and cause the angle and speed of the corresponding obstacle avoidance output to jump)

Images