System for learning control commands to robotically move a load, especially suitable for use in cranes to reduce load sway

The electronic anti-sway system involves two modes, a "LEARN mode" and an "AUTO mode". In the LEARN mode, an experienced operator operates the crane manually while his specific control movements are observed by the inventive system. The movements are stored, along with such parameters as load position as a function of time, and the weight of the load. Preferably, for loads and movement paths which are substantially identical, only the most efficient path produced by the experienced human operator is recorded permanently, less efficient paths being discarded. A library of preferred paths is thus accumulated, preferably with one preferred path for each type of load and source/destination. Thereafter, in the "AUTO mode", an operator may entrust movement of the load to the present system, which causes the load to efficiently and safely traverse an optimum path (with minimum sway) in a minimum period of time. Preferably, various safeguards are provided by the system. For example, the crane is preferably manually controlled during the very beginning and end portions of the load's movement. Further, if the path traversed by a load in the "AUTO" mode deviates significantly from the projected paths recorded in the library, the system automatically stops the load's movement and surrenders control to the human operator.