Coordinating use of different motion prediction models to predict a location of a mobile robot at a future point in time

A method for coordinating use of different motion prediction models, comprising:

• determining, by a processor operating a first motion prediction model, a first prediction of a location of a mobile robot at a future point in time;
• determining, by the processor operating a second motion prediction model, a second prediction of the location;
• causing, by the processor and based on a determination of an existence of a condition, a control signal to be sent to the mobile robot to move, based on at least the first prediction, to prevent an indeterminant value being used as the location, the first motion prediction model being capable, during the existence of the condition, of producing an indeterminant value for the first prediction, the condition being a speed of the mobile robot being less than a threshold speed or a degree of curvature of a road being greater than a threshold degree;
• causing, by the processor, a cessation of an operation of the first motion prediction model in response to the determination of the existence of the condition for greater than a predetermined threshold duration of time;
• and
• moving, in response to a receipt of the control signal, the mobile robot.

A system for coordinating use of different motion prediction models, comprising:

• a processor;
• and
• a memory storing:
• a first motion prediction model module including instructions that cause the processor to:
• determine, operating a first motion prediction model, a first prediction of a location of a mobile robot at a future point in time;
• and
• cause a cessation of an operation of the first motion prediction model in response to a determination of an existence of a condition for greater than a predetermined threshold duration of time, the condition being a speed of the mobile robot being less than a threshold speed or a degree of curvature of a road being greater than a threshold degree;
• a second motion prediction model module including instructions that cause the processor to determine, operating a second motion prediction model, a second prediction of the location;
• and
• an actuator module including instructions that cause, based on the determination of the existence of the condition, the mobile robot to move, based on at least the second prediction to prevent an indeterminant value being used as the location, by changing at least one of a direction or a speed, the first motion prediction model being capable, during the existence of the condition, of producing an indeterminant value for the second prediction.