An AI-assisted mobile tennis training system integrates a motorized ball delivery device, multi-camera vision, autonomous navigation, and artificial intelligence for adaptive, data-driven player development. The device includes a motorized chassis, ball hopper, programmable ball delivery mechanism with variable spin, speed, and trajectory control, and a navigation subsystem employing omnidirectional wheels and localization based on court line recognition. A multi-camera array captures real-time player and ball movement, while onboard and remote computing modules process the data to determine player position, shot type, and performance metrics. An AI model predicts optimal ball delivery parameters and adapts drills based on player progress. A mobile application provides remote control, drill customization, and performance analytics. The system supports autonomous repositioning, safety monitoring, and individualized training plans, enabling dynamic, responsive, and efficient tennis practice for skill acquisition and improvement in both amateur and professional players.

A work management system includes: a working robot configured to perform work while autonomously traveling on a field; a management facility configured to manage the field and balls; and a management device configured to know a management situation of the balls. A work schedule of the working robot is adjusted depending on a ball management situation known by the management device.

A method, apparatus, and system are disclosed for a robot with tennis ball gathering capabilities. A robot is configured to collect, transport, and deposit for storage tennis balls and other light-mobile sports equipment such as are used in table tennis, badminton, squash, pickleball, golf, basketball, dodgeball, floor hockey, indoor soccer, etc., to assist a player in practice without need for additional support personnel or manual ball retrieval by the player.