Linear Temporal Logic based Motion Planning

We describe the problem of motion planning for multiple goals in unstructured environments by using temporal logic for specification of the goals. We solve the problem of mission planning for mobile robots, wherein the goal of the robot can be a complex mission specification rather than a classical "Go from A to B" problem. To achieve this problem we take a two dimensional map of the real environment and identify the regions of interest, navigable areas and non-navigable areas, which are tagged to make the final map. During the motion of robot current position in stationary environment is also known to us. We describe the high level planning algorithm which gives a trajectory catering to the temporal logic based goal specification. We use optimal graph search to search for an optimal solution based on complex goal specifications, while at each step checking for the validity of the temporal logic. Simulation results to various practical scenarios are provided making a robot act over high level specifications while yielding optimal plans. We describe the linear temporal logic-based reactive motion planning. The desire task specification may consist of behaviors of robot, environment constraints, task optimality like obstacle avoidance, rescue, surveillance, safety etc. In another approach we use LTLMoP, ROS based environment for our computational approach start with design such real-time inspired complex environment then using Linear Temporal Logic (LTL) motion goal specification generated also validate the feasibility of specification.Then finally controller is used to generate desire trajectory to achieve such goal for mobile robot. However during the simulation experiments of our experimental lab environment, we use LTLMoP, ROS based environment. Moreover during the preparation of simulation result we use PR2 Robot, Basic SIM mobile robot and Point Robot for navigation based on goal specifications.Our all simulation results generated for high level controllers, low level planners may fetch simultaneously for path planning and controlling the physical behavior of robot.