Applications of Artificial Intelligence in Safe Human Robot Interactions

    28 Votes

This presentation discuss about the Integration of both Robots and Humans workspace. A new sensory system for modelling,tracking and predicting human motions within a Robot workspace is introduced. Our objective is to obtain a super quadric based model of human using SOM. Also assess the danger of the robot operations. A new reactive control scheme is introduced.

Human Modelling

Human Modelling for Person De-Identification

Four steps in Human modelling are 

  • The safety mat consists of a number of pressure activated nodes.Each node on the mat has fixed coordinates. Under the human body weight, a set of nodes F = {(xj, yj)|j = 1, . . . , n} are activated across the mat. 
  • This set is then clustered into two subsets F1 and F2, corresponding to each foot using a SOM network .
  • Using these subsets, the human body orientation and its location are derived .
  • This information along with average human body dimensions is then used in order to obtain a model of the human .

Safety mat detects obstacles. It is constructed using 2 rubber sheets having parallel wires. It contains pressure activated nodes. Each node on the mat has fixed coordinates.

Data processing using SOM

Data processing using SOM

In Data processing using SOM, activated node set F needs to be first divided(clustered) into two subsets,F1 and F2 corresponding to each foot. SOM network seems a suitable candidate for clustering the data representing  human footprints. Input to the SOM network is (xj, yj) pairs and produce output as (f1,f2). Type A sample set have 2 zero eigen values in its Laplacian metrics. To convert type B to type A, uncertain nodes are need to be removed. L1 and L2 corresponding to the outer borders and orientation of each soles. L avg represents inner border of the two soles. Body orientation can be obtained from 2 subsets F1 & F2.

Obtaining body orientation

  • alpha - average of sole orientation
  • Centre of the body
  • These values used to obtain human model
  • Lines lL and lR lines connecting the centers of the forefoot to the heel in each sole, respectively.

Prediction of the human trajectory is motivated by ordinary human-human interaction. Observe the pattern of the motion and predict the motion using ANN.

Conclusion

Study on a sensory system and reactive control scheme. SOM network and super quadric functions are used for human modelling. Human motion predicted using ANN. Reactive control scheme is developed.

Attachments:
Download this file (Applications of Artificial Intelligence in Safe Human-Robot Interactions.ppt)Artificial Intelligence in Safe Human Robot Interactions[PPT Presentation]3130 Kb