Non-smooth environment modeling and global path planning for mobile robots

An Approximate Voronoi Boundary Network is constructed as the environmental model by way of enlarging the obstacle raster. The connectivity of the path network under complex environment is ensured through building the second order Approximate Voronoi Boundary Network after adding virtual obstacles at joint-close grids. This method embodies the network structure of the free area of environment with less nodes, so the complexity of path planning problem is reduced largely. An optimized path for mobile robot under complex environment is obtained through the Genetic Algorithm based on the elitist rule and re-optimized by using the path-tightening method. Since the elitist one has the only authority of crossover, the management of one group becomes simple, which makes for obtaining the optimized path quickly. The Approximate Voronoi Boundary Network has a good tolerance to the imprecise a priori information and the noises of sensors under complex environment. Especially it is robust in dealing with the local

Non-smooth environment modeling and global path planning for mobile robots

An Approximate Voronoi Boundary Network is constructed as the environmental model by way of enlarging the obstacle raster. The connectivity of the path network under complex environment is ensured through building the second order Approximate Voronoi Boundary Network after adding virtual obstacles at joint-close grids. This method embodies the network structure of the free area of environment with less nodes, so the complexity of path planning problem is reduced largely. An optimized path for mobile robot under complex environment is obtained through the Genetic Algorithm based on the elitist rule and re-optimized by using the path-tightening method. Since the elitist one has the only authority of crossover, the management of one group becomes simple, which makes for obtaining the optimized path quickly. The Approximate Voronoi Boundary Network has a good tolerance to the imprecise a priori information and the noises of sensors under complex environment. Especially it is robust in dealing with the local or partial changes, so a small quantity of dynamic obstacles is difficult to alter the overall character of its connectivity, which means that it can also be adopted in dynamic environment by fusing the local path planning.