Table 2.

Path planning methods for industrial robots.

Path planning methods	Representative algorithms	Advantages	Disadvantages
Traditional obstacle avoidance planning methods	APF (artificial potential field method) (Khatib, 1986), A* (Hart et al., 1968)	The principle is simple and easy to implement	The methods need a large amount of calculation, are easy to fall into local minimum value, and cannot be applied to higher-dimensional space path planning
Intelligent obstacle avoidance planning methods	Artificial neural network (Wang et al., 2009), ant colonies algorithm (Guan-Zheng et al., 2007)	Easy to implement without modeling the environment	The methods have randomness, the solution is not unique, and they are not suitable for high-dimensional spatial path planning
Obstacle avoidance planning method based on random sampling	PRM (Probalistic Roadmaps) (Kavraki et al., 1996), RRT (Rapidly-exploring Random Trees) (LaValle, 1998)	It does not depend on the robot's state space and is suitable for high-dimensional space path planning	The methods have randomness and a slow search speed

The symbol * represents a special flag for an improved RRT algorithm, which is a specific sign.