Path planning using heuristic algorithm in dynamic environment

Navigation of mobile robots and autonomous vehicles is one of the important issues in computer and control sciences. Path planning and obstacle avoidance are current topics of navigational challenges for mobile robots and autonomous vehicles. One of the essential issues in robotics is path planning. The problem of path planning is a well-known NP-hard problem where the complexity increases with the degrees of freedom of the vehicle or robot. The main aim of path planning is to find a safe and smooth path in a dangerous environment for a mobile robot or vehicle so that the robot moves from the starting point to the destination point without colliding with obstacles. This thesis has investigated path planning issues for navigation of a mobile robot and an autonomous vehicle with two different approaches. First, a novel path planning approach using the grasshopper algorithm is presented for the navigation of a mobile robot in dynamic and unknown environments. For this purpose, two different approaches are presented. A sensory system is used to detect the obstacles and a new method was developed to predict and avoid static and dynamic obstacles while the velocities of the obstacles are unknown. The mobile robot uses the obtained information and finds a collision-free, optimal and safe path. The proposed approach in this thesis was tested in crowded and complex environments. Simulation results demonstrate that the approach is successful in all test environments. Also, the proposed approach is compared with several heuristic methods and hybrid approaches. The comparison work stipulates that the approach introduced here is promising in terms of running time, optimality, stability and failure rate. Second, a new path planning based on Model Predictive Controller (MPC) for an autonomous vehicle is developed, which automatically decides about the mode of maneuvers such as lane-keeping, lane changing, and lane merging. For ensuring safety, we have additionally used three different potential field functions for road boundary, obstacles, and center of the lane. The potential fields of road boundaries keep the vehicle from going out of the road boundaries, the potential field surrounding vehicle keeps the vehicle away from obstacles or surrounding vehicles and also the potential field of lane centering leads to tracking of the centerline of the lane by the autonomous vehicle. The proposed path planning controller on the different scenarios have been tested. The obtained results represent by using a path planning controller the vehicle avoids collision with obstacles and observes the regulations of the road by appropriate dynamics of the vehicle. The path planning controller guarantees the safety of the autonomous vehicle.