Bi-Directional Adaptive Probabilistic Method With a Triangular Segmented Interpolation for Robot Path Planning in Complex Dynamic-Environments

Path planning is a fundamental aspect of mobile robots and autonomous systems. Methods of path planning are used in robotics to create a path for a robot or autonomous system to follow from a starting position to a goal one while avoiding obstacles and satisfying any additional conditions. There are many different methods to plan the path, including probabilistic methods, heuristics-based approaches, and optimization-based methods. In this paper, we introduce a novel path planning method called Dynamic Adaptive RRT-connect with Triangular Segmented Interpolation. Our approach aims to enhance the conventional Rapidly-exploring Random Tree (RRT) algorithms by incorporating an Adaptive-RRT strategy. This strategy involves selecting a random node as a new node to augment the exploration of the tree, thereby improving its coverage of the search space. Furthermore, we employ a Bi-directional scheme to further enhance the convergence time and cost of our method. By exploring the search space from both the initial and goal configurations simultaneously, we exploit the advantages of a two-way search, potentially resulting in more efficient and optimized paths. To improve the quality of the generated paths, our method leverages the Triangular Segmented Interpolation (TSI) technique. TSI helps in reducing the path length and increasing its smoothness by interpolating between the configurations in a triangular segmented manner, resulting in more natural and feasible trajectories. Moreover, considering the dynamic nature of the environment, our method operates within the framework of the Dynamic Window Approach (DWA). By adapting to the changing environment, our approach effectively avoids dynamic obstacles and navigates the robot or system through complex and unpredictable scenarios. We have conducted extensive experiments in various environments to evaluate the performance of our proposed method. The results demonstrate that our approach outperforms the individual RRT and RRT-connect algorithms in terms of computation time (reduced by 90-80%), cost (reduced by 82-63%), and path length (shortened by 17-12%). Additionally, our method exhibits efficient obstacle avoidance capabilities, enabling successful navigation in dynamic environments.