The Research Of The Mobile Robot Path Planning

The research and development of mobile robots have been paid great attention to in recent years. The requirement of its ability to take corresponding measures in accordance with the environmental changes in its moving courses has become higher and higher. Therefore, the path planning, as an important field in robot intelligence, appears particularly significant. Path planning is searching an optimal path without collision from the start to the end under a certain index of performance.This thesis does the research on the path planning problem for mobile robot in static known environment and static unknown environment.The main content of this thesis is given as follows.For the path planning in static known environment, a new method based on the parameterized optimization is proposed. The path planning problem is formulated as an optimal control problem with constraints.Next, the optimal control problem is transformed into the static nonlinear programming problem with parameterized optimization method, and a new augmented-SWIFT(Sequential Weight-Increasing Factor Technique) is also presented. The SWIFT is used in static constrained optimization problem, whereas the new augmented-SWIFT is extended to calculate dynamic constrained optimal control problem. Finally, the feasibility of this algorithm is verified by simulation.Another optimal control based path planning method is discussed by using the gauss pseudospectral approximation. All the state and control variables are approximated by the Lagrange interpolating polynomials. Then the continuous optimal control problem is discretized and transcribed to a nonlinear programming problem (NLP).which can be solved by using the sequential quadratic programming (SQP) method. This method has no request on the initial values.For the static unknown environment, the Khatib artificial potential field model is analysed and the problem of goals non-reachable with obstacles nearby (GNRON) when using the traditional potential field methods for mobile robot path planning is described. In order to avoid the path planning problem of GNRON, the relative positions of the robot and the goal is taken into considering to improve the artificial potential field repulsive function. Simulation results are given, and show that no passage between closely space obstacles in some cases. To improve the shortcoming of the preceding method, another new method based on the circle scanning is introduced. Instead of changing the potential field function of the repulsive force in traditional APF methods, the improved APF applies the circle scanning method combining with the potential field function to search the optimal point in the path. This method can avoid collision with obstacles and reach the goal. The simulation result illustrates the effectiveness of the proposed method.