Robot Path Planning _ Artificial Potential field method _ machine learning

Robot Path Planning _ Artificial Potential field method

Principle

Artificial potential field method is a kind of virtual force method proposed by Khatib. The principle is: the movement of robots in the environment as a robot in the virtual Artificial force field movement. Obstacles to the robot to create repulsion, the target point to the robot to generate gravity, gravity and repulsion force as the acceleration of the robot to control the robot's direction and calculate the location of the robot.

The gravitational field (attraction) is monotonically increasing with the distance between the robot and the target point, and the direction points to the target point.

Repulsion field (repulsion)

When the robot is in the obstacle position, it has a maximal value, and decreases with the distance between the robot and the obstacle, and the direction points away from the obstacle direction.

Advantages

Simple and practical, good real-time performance

The structure is simple and convenient for real-time control of the bottom, and it is widely used in real-time obstacle avoidance and smooth trajectory control. Disadvantages

1 There is a trap area

2 The path is not recognized in a similar group of obstacles

3 Shock before the obstruction

4) Swinging in a narrow passage

5 The object near the obstacle can not be reached

The range of gravitational potential field is large, and the scope of repulsion is only local, when the distance between the robot and the obstacle exceeds the range of the obstacle, the robot is not affected by the repulsion potential field. Therefore, the potential field method can only solve the local space obstacle avoidance problem, it lacks the global information, so it is easy to fall into the local minimum value. The local minimum value point is the joint distribution of the gravitational potential field function and the repulsive potential field function, and in some regions, it is affected by several functions, which results in the local minimum point. When the robot is in the local minimum point, the robot is prone to oscillation or stagnation. The more obstacles, the greater the likelihood of local minima, and the more the number of local minima will be produced.