Safe, efficient waypoint manipulation for path planning of non-holonomic robots
Akhilesh Bhat, Natsuki Kai, Takayuki Suzuki, Takahiro Shiroshima, Hiroshi Yoshida · 2022 27th International Conference on Automation and Computing (ICAC) · 2022
During the pandemic, health and safety concerns kept people indoors and increased the demand for same-day delivery. This put a huge stress on the e-commerce industry. Adoption of warehouse automation skyrocketed, leading to an increased emphasis on efficiency. In warehouses where large industrial robots like forklifts are deployed, efficiency and safety, both are important. Conventional methods like Reeds-Shepp (RS) that consider the motion model are often sub-optimal as they do not consider obstacles, while those that are optimal like Dijkstra and A*are difficult to implement due to kinematic constraints. This paper proposes a hierarchical structure consisting of an A*-based global planner which goes through a pre-processing step. The local planner then generates a Reeds-Shepp curve to a waypoint chosen using a ”cost” parameter.