Complete coverage path planning for reconfigurable omni-directional mobile robots with varying width using GBNN(n)

Self-reconfigurable robots which can change their footprint's width have demonstrated the ability to access confined areas. To enhance the efficiency of area coverage in complete coverage path planning (CCPP) for complex and confined environments, it is advantageous to cover areas fast in wide areas while maintaining the ability to navigate through tight spaces and cover hard to access areas. This can be achieved by leveraging the flexibility of an omnidirectional self-reconfigurable robot footprint. The widest footprint is used to speed up the area coverage when there are no obstacles around, while the smallest footprint is used to navigate through tight spaces. However, the generation of robotic width reconfiguration state during autonomous CCPP generation, poses challenges. In this paper, a CCPP for omni-directional robots of varying width with n-reconfiguration states is proposed. To this end, the proposed CCPP is a modified GBNN with n-reconfiguration states (GBNN(n)). It generates the global path autonomously, which determines the robot width as per the nth reconfiguration states so as to increase area coverage in open areas and reduce robot footprint in tight spaces. The proposed complete coverage path planning are compared against state-of-the-art GBNN and CCPP optimization using depth-limited search and successfully demonstrate that the proposed algorithm helps robots of varying widths achieve higher area coverage in lesser steps, energy and distance. The supporting simulation and experimental video link1 is also provided to highlight the outcomes.

Automated summary

To enhance the efficiency of area coverage in complex and confined environments, a complete coverage path planning algorithm for omnidirectional self-reconfigurable robots with varying width is proposed. The algorithm leverages the flexibility of the robot's variable width to cover wide areas quickly and navigate through tight spaces. It generates a global path that determines the robot's width to increase area coverage in open areas and reduce the footprint in tight spaces.