An exercise machine for assisting a user in targeting a muscle for an exercise motion. A sensor detects a current use condition while the user performs the exercise motion. The current use condition is one of multiple use conditions. A current activation level of the muscle changes within the multiple use conditions. A control system is operatively coupled to the sensor and is configured to determine the current activation level of the muscle based on the current use condition detected by the sensor, receive a desired activation level of the muscle, and determine a recommended use condition within the multiple use condition to cause the muscle to be at the desired activation level for performing the exercise motion, and to notify the user of the recommended use condition and/or automatically cause the exercise machine based on the recommended use condition, respectively, to assist the user in targeting the muscle.

A system preparing kinematic recommendations for athletes using artificial intelligence is disclosed. The system is programmed to sense personal physical data regarding a subject athlete, and optionally review environmental data, and historical performance and physical data related to the subject athlete. The system produces performance recommendations for that subject athlete, determines a best medium for providing the performance recommendations to the subject athlete based on their personal physical data, and provides the performance recommendations to the subject athlete via that determined best medium. The best medium can include an augmented reality interface worn by the athlete. The performance recommendations can be prepared by machine learning or artificial intelligence models designed to analyze athletic performances, and to communicate conversationally with the athlete.

A method for controlling an exercise machine to assist a user in targeting a muscle for performing an exercise motion. The method includes receiving a target muscle as the muscle to target when performing the exercise motion. The method further includes determining, from a plurality of use conditions, a recommended use condition that would cause operating the exercise machine to focus on the target muscle received. The method further includes guiding the user based on the recommended use condition and/or automatically causing the exercise machine to operate based on the recommended use condition to thereby assist the user in targeting the muscle for performing the exercise motion.

The invention concerns a training accessory for an electronic device and a training method configured to provide training feed-back in real-time. The training accessory comprises a housing suitable for holding the electronic device (400) having a graphical user interface (401), a data interface between the accessory (1) and the electronic device (400), one or more elastically deformable holding means (20), each of said holding means being configured to freely swivel around its attachment point, and one or more sensors (3), including at least a force sensor in the accessory (1), the force sensor being configured to determine the force exerted by the holding means (20). The method includes the steps of determining or receiving a safe training range of an individual user, receiving user training data pertaining to the user's execution of a training exercise in real-time, and based on said user training data calculating corrections in real-time and communicating instructions to the user in real-time to permit the user to correct his execution of the exercise such that the exercise is executed within a safe training range.