Bootstrap Learning and Visual Processing Management on Mobile Robots

Algorithm 2

Motion sequence generation.

Require: Ability to learn color models.

Require: Positions, shapes and color labels of the objects of

  interest in the robot's environment. Initial robot pose.

(1) Move between randomly selected target poses.

(2)  CollectMEMData() – collect data for motion error model.

(3)  CollectColLearnStats() – collect color learning statistics.

(4)  NNetTrain() – Train the Neural network for the MEM, (8).

(5)  UpdateFM() – Generate the statistical feasibility model, (9).

(6)  GenCandidateSeq() – Generate candidate sequences, (10).

(7)  EvalCandidateSeq() – Evaluate candidate sequences.

(8)  SelectMotionSeq() – Select final motion sequence.

(9) Execute motion sequence and learn colors – Algorithm

  described in [76].