Robotic disassembly re-planning using a two-pointer detection strategy and a super-fast bees algorithm

Automated disassembly of End-of-Life (EoL) products can be difficult to implement due to uncertainties in their conditions. An automatic re-planning function is required to enable flexible adjustments of disassembly plans and thus increase disassembly efficiency. The re-planning function is able to detect subassemblies and separable components, and adjust disassembly sequences and directions when components interlock and are irremovable. This paper presents a two-pointer detection strategy to find detachable subassemblies very quickly. A summation operator and a list with two pointers are used to check the interferences between components in a minimum number of steps. Then, a ternary bees algorithm is proposed to identify new disassembly sequences and directions. The algorithm combines the merits of a greedy search and meta-heuristic techniques by using only three collaborative potential solutions and three concurrent operations. Experimental results show that the proposed approach is able to perform a rapid subassembly detection and sequence optimisation for a robotic disassembly task, thus allowing real-time re-planning.