Experiment 4 (Stop Behavior)

The fourth experiment of robot behavior, shown in Figure 4.6, is to test the Stop behavior. A stop would happen if there is an urgent or unexpected situation. Also, by calculating the cost, a stop in a short time may overcome some situations.

(a) (b)

(c) (d) Figure 4.7. Illustration of Stop Behavior.

In Figure 4.6 (a) and (b) above, the robot follows the two pedestrians who are belongs to a same group. Because of the interacting state of this group, which is interacting, the robot would maintain a lager distance from them. When another pedestrian (on the left) tries to cross by, shown in (c) and (d), robot use the Stop behavior to yield to the person and maintain physical safety, successfully.

4.6 Evaluation and Discussion

To evaluate and validate our system, we define two metrics that quantify the different aspects of the navigational behavior of the robot. These two metrics are the measurements of the cost mentioned in Section 2.5.

 Progress:

Measurement of distance in each time frame. A larger distance would perform a worse score of this item.

 Force:

The summarization of maximum repulsive force from other agents which penalizes the robot if it is close to encounters.

We use these metrics to compare our method with a non-holonomic mobile robot applying the traditional SFM. The same scenarios of experiment 2 to 4 are used and the results are shown below. The experimental data collects 15 times of experiments and the results are the mean values. All the measurements are normalized based on the corresponding non-holonomic results.

(a) (b)

(c)

Figure 4.8. Evaluation Metrics of Experiments.

By the figures above, the differences of these two system are shown. In the experiment 2, there are obvious difference in both Progress and Force. From the experiment, the behaviors which a non-holonomic mobile system executed are Go-Forward, Agent Following, and Stop. Because of the complexity of indoor environment and the mobility of the non-holonomic robot, the Stop behavior is executed much more frequently than our system which causes the score of Progress getting higher. On the other hand, the score of Force is also higher than ours in the reason that the solution space of executable velocity in our system is much wider. A smaller cost of behavior can be executed by the holonomic robot.

In the experiment 3, the environment, the corridor, is simpler. Both of compared

method and our system mainly execute the Wall Following behavior. In this case, the differences of evaluation metrics come from the rotational behavior and interacting factor effect. In the experiment, when robot encounters the group of opposite pedestrians, it performs a movement to avoid people. A diagonal movement with rotation is executed by our system and a turning movement is executed by the non-holonomic mobile robot. The difference of turning behavior causes the difference in the Progress metric. For the Force metric, since we introduce the interacting factor, the social space of crowds is considered.

The repulsive forces from agents in our method perform a lower scale when compare to another method in the same distance.

In the experiment 4, the differences in both metrics are not so obvious. In this case, a Stop and the Group Following behavior are executed. The difference of Progress is also from the turning behavior which robot uses it to adjust its position. This turning behavior is part of the Go-Forward behavior which the moving direction is different from robot’s orientation. The difference of Force comes from the interacting factor effect. Our system would maintain a farther distance compared to non-holonomic one.

Chapter 5

Conclusion

In this thesis, a socially-aware omnidirectional mobile robot navigation is proposed.

By the sociological researches, human motion and crowd have been studied for decades.

Based on these researches, we construct a grouping model that would perform social space during walking and develop several potential field-based robot behaviors to conquer different situations. Applying to an omnidirectional mobile robot, the holonomic motion can be achieved. For a robot navigate besides human, it should perform psychological safety and move toward its goal.

We start at dividing pedestrians into groups that the pedestrian behaviors are similar inside the same group. Human sometimes tend to walk and interact with others during navigation. To respect to the social space during walking, we define an interacting factor that can perform the interacting state of group. By the fact that human is able to navigate in one direction and face to another direction simultaneously, we apply an omnidirectional mobile robot to perform a natural movement and a high mobility. Then based on the holonomic kinematic, we proposed a rotational social force that describes the orientation of robot during the navigation. This modified Social Force Model, extended Social Force Model, is designed. Based on the extended model, we design several robot behaviors in different functionality, which are Go-Forward, Agent/Group Following, Wall Following and Stop. By applying the Multi-Policy Decision Making procedure, the robot navigation framework is completed. Finally, our approach has been run on several experiments that show the grouping procedure, robot behaviors and total navigation performance which

illustrate the feasibility and mobility of our socially-aware omnidirectional mobile robot navigation.

REFERENCE

[1] Dirk Helbing and Péter Molnár. Social force model for pedestrian dynamics. Phys.

Rev. E 51, 1988.

[2] Thibault Kruse, Amit Kumar Pandey, Rachid Alami, and Alexandra Kirsch.

Human-aware robot navigation: A survey. Robotics and Autonomous Systems, 61(12):1726 – 1743, 2013.

[3] Edward Twitchell Hall. The hidden dimension. 1966.

[4] Pacchierotti, Elena & Christensen, Henrik & Jensfelt, Patric. (2005). Embodied Social Interaction for Service Robots in Hallway Environments. 25. 293-304.

10.1007/978-3-540-33453-8_25.

[5] C. P. Lam, C. T. Chou, K. H. Chiang and L. C. Fu, "Human-Centered Robot Navigation—Towards a Harmoniously Human–Robot Coexisting Environment,"

in IEEE Transactions on Robotics, vol. 27, no. 1, pp. 99-112, Feb. 2011.

[6] Javier Minguez and L. Montano, "Nearness diagram (ND) navigation: collision avoidance in troublesome scenarios," in IEEE Transactions on Robotics and Automation, vol. 20, no. 1, pp. 45-59, Feb. 2004.

[7] Kessler J., Schroeter C., Gross HM. (2011) Approaching a Person in a Socially Acceptable Manner Using a Fast Marching Planner. In: Jeschke S., Liu H., Schilberg D. (eds) Intelligent Robotics and Applications. ICIRA 2011.

[8] M. Luber, L. Spinello, J. Silva and K. O. Arras, "Socially-aware robot navigation:

A learning approach," 2012 IEEE/RSJ International Conference on Intelligent Robots and Systems, Vilamoura, 2012, pp. 902-907.

[9] W. Chi, H. Kono, Y. Tamura, A. Yamashita, H. Asama and M. Q. H. Meng, "A human-friendly robot navigation algorithm using the risk-RRT approach," 2016

IEEE International Conference on Real-time Computing and Robotics (RCAR), Angkor Wat, 2016, pp. 227-232.

[10] A. Bellarbi, S. Kahlouche, N. Achour and N. Ouadah, "A social planning and navigation for tour-guide robot in human environment," 2016 8th International Conference on Modelling, Identification and Control (ICMIC), Algiers, 2016, pp.

622-627.

[11] A. K. Pandey and R. Alami, "A framework towards a socially aware Mobile Robot motion in Human-Centered dynamic environment," 2010 IEEE/RSJ International Conference on Intelligent Robots and Systems, Taipei, 2010, pp. 5855-5860.

[12] S. B. Liu, H. Roehm, C. Heinzemann, I. Lütkebohle, J. Oehlerking and M. Althoff,

"Provably safe motion of mobile robots in human environments," 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Vancouver, BC, 2017, pp. 1351-1357.

[13] Y. Tamura, T. Fukuzawa and H. Asama, "Smooth collision avoidance in human-robot coexisting environment," 2010 IEEE/RSJ International Conference on Intelligent Robots and Systems, Taipei, 2010, pp. 3887-3892.

[14] Zanlungo, Francesco & Ikeda, Tetsushi & Kanda, Takayuki. (2011). Social force model with explicit collision prediction. EPL (Europhysics Letters). 93. 68005.

10.1209/0295-5075/93/68005.

[15] G. Ferrer, A. Garrell and A. Sanfeliu, "Robot companion: A social-force based approach with human awareness-navigation in crowded environments," 2013 IEEE/RSJ International Conference on Intelligent Robots and Systems, Tokyo, 2013, pp. 1688-1694.

[16] D. Katagami and S. Yamada, “Active teaching for an interactive learning robot,” in

[17] G. Ferrer and A. Sanfeliu, "Behavior estimation for a complete framework for human motion prediction in crowded environments," 2014 IEEE International Conference on Robotics and Automation (ICRA), Hong Kong, 2014, pp. 5940-5945.

[18] G. Ferrer and A. Sanfeliu, "Proactive kinodynamic planning using the Extended Social Force Model and human motion prediction in urban environments," 2014 IEEE/RSJ International Conference on Intelligent Robots and Systems, Chicago, IL, 2014, pp. 1730-1735.

[19] B. S. B. Dewantara, "Building a socially acceptable navigation and behavior of a mobile robot using Q-learning," 2016 International Conference on Knowledge Creation and Intelligent Computing (KCIC), Manado, 2016, pp. 88-93.

[20] K. Mombaur, J. P. Laumond and E. Yoshida, "An optimal control model unifying holonomic and nonholonomic walking," Humanoids 2008 - 8th IEEE-RAS International Conference on Humanoid Robots, Daejeon, 2008, pp. 646-653.

[21] Hughes, Rowan & Ondrej, Jan & Dingliana, John. (2014). Holonomic Collision Avoidance for Virtual Crowds. 10.2312/sca.20141128.

[22] Tan-Viet-Anh Truong. Unifying nonholonomic and holonomic behaviors in human locomotion. Automatic. Institut National Polytechnique de Toulouse - INPT, 2010.

[23] S. A. Yang, E. Gamborino, C. T. Yang and L. C. Fu, "A study on the social acceptance of a robot in a multi-human interaction using an F-formation based motion model," 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Vancouver, BC, 2017, pp. 2766-2771.

[24] A. Khelloufi, N. Achour, R. Passama and A. Cherubini, "Tentacle-based moving obstacle avoidance for omnidirectional robots with visibility constraints," 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Vancouver, BC, 2017, pp. 1331-1336.

[25] Vinciarelli, A., Pantic, M., Bourlard, H., & Pentland, A. (2008). Social signal processing: state-of-the-art and future perspectives of an emerging domain. ACM Multimedia.

[26] J. Mumm and B. Mutlu, "Human-robot proxemics: Physical and psychological distancing in human-robot interaction," 2011 6th ACM/IEEE International Conference on Human-Robot Interaction (HRI), Lausanne, 2011, pp. 331-338.

[27] A. Gaschler, S. Jentzsch, M. Giuliani, K. Huth, J. de Ruiter and A. Knoll, "Social behavior recognition using body posture and head pose for human-robot interaction," 2012 IEEE/RSJ International Conference on Intelligent Robots and Systems, Vilamoura, 2012, pp. 2128-2133.

[28] Rios-Martinez, Jorge & Spalanzani, Anne & Laugier, Christian. (2014). From Proxemics Theory to Socially-Aware Navigation: A Survey. International Journal of Social Robotics. 7. 137-153. 10.1007/s12369-014-0251-1.

[29] P. Althaus, H. Ishiguro, T. Kanda, T. Miyashita and H. I. Christensen, "Navigation for human-robot interaction tasks," Robotics and Automation, 2004. Proceedings.

ICRA '04. 2004 IEEE International Conference on, 2004, pp. 1894-1900 Vol.2.

[30] J.T. Butler, A. Agah, Psychological effects of behavior patterns of a mobile personal robot, Autonomous Robots 10 (2) (2001) 185–202.

[31] Dautenhahn, K., Walters, M.L., Woods, S.N., Koay, K.L., Nehaniv, C.L., Sisbot, E.A., Alami, R., & Siméon, T. (2006). How may I serve you? a robot companion approaching a seated person in a helping context. HRI.

[32] Koay, Kheng & Sisbot, Emrah & Syrdal, Dag Sverre & Walters, Michael &

Dautenhahn, Kerstin & Alami, Rachid. (2007). Exploratory Study of a Robot Approaching a Person in the Context of Handing Over an Object. AAAI Spring

[33] Y. F. Chen, M. Everett, M. Liu and J. P. How, "Socially aware motion planning with deep reinforcement learning," 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Vancouver, BC, 2017, pp. 1343-1350.

[34] Harmish Khambhaita, Jorge Rios-Martinez, Rachid Alami. Head-Body Motion Coordination for Human Aware Robot Navigation. 9th International workshop on Human-Friendlly Robotics (HFR 2016), Sep 2016, Génes, Italy. 8p., 2016.

[35] T. Matthew Ciolek and Adam Kendon. Environment and the Spatial Arrangement of Conversational Encounters. Sociological Inquiry, 50(3-4):237–271, 1980.

[36] J. Rios-Martinez, A. Spalanzani and C. Laugier, "Understanding human interaction for probabilistic autonomous navigation using Risk-RRT approach," 2011 IEEE/RSJ International Conference on Intelligent Robots and Systems, San Francisco, CA, 2011, pp. 2014-2019.

[37] A. Vega-Magro, L. Manso, P. Bustos, P. Núñez and D. G. Macharet, "Socially acceptable robot navigation over groups of people," 2017 26th IEEE International Symposium on Robot and Human Interactive Communication (RO-MAN), Lisbon, 2017, pp. 1182-1187.

[38] Truong, Xuan-Tung & Ngo, Trung Dung. (2016). Dynamic Social Zone based Mobile Robot Navigation for Human Comfortable Safety in Social Environments.

International Journal of Social Robotics. 8. 10.1007/s12369-016-0352-0.

[39] K. O. Arras, O. M. Mozos and W. Burgard, "Using Boosted Features for the Detection of People in 2D Range Data," Proceedings 2007 IEEE International Conference on Robotics and Automation, Roma, 2007, pp. 3402-3407.

[40] J. Xavier, M. Pacheco, D. Castro, A. Ruano and U. Nunes, "Fast Line, Arc/Circle and Leg Detection from Laser Scan Data in a Player Driver," Proceedings of the 2005 IEEE International Conference on Robotics and Automation, 2005, pp.

3930-3935.

[41] M. S. Arulampalam, S. Maskell, N. Gordon, and T. Clapp. A tutorial on particle filters for online nonlinear/non-gaussian bayesian tracking. IEEE Transactions on Signal Processing, 50(2):174–188, Feb 2002.

[42] Z. Cao, T. Simon, S. E. Wei and Y. Sheikh, "Realtime Multi-person 2D Pose Estimation Using Part Affinity Fields," 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, 2017, pp. 1302-1310.

[43] K. Simonyan and A. Zisserman. Very deep convolutional networks for large-scale image recognition. In ICLR, 2015

[44] D. Mehta, G. Ferrer and E. Olson, "Autonomous navigation in dynamic social environments using Multi-Policy Decision Making," 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Daejeon, 2016, pp. 1190-1197.

[45] Buchmueller, S., Weidmann, U. (2006): Parameters of Pedestrians, Pedestrian Traffic and Walking Facilities, IVT-Report Nr. 132, Institut for Transport Planning and Systems (IVT), Swiss Federal Institute of Technology Zurich (ETHZ)

[46] Moussaïd M, Perozo N, Garnier S, Helbing D, Theraulaz G (2010) The Walking Behaviour of Pedestrian Social Groups and Its Impact on Crowd Dynamics. PLOS ONE 5(4): e10047

[47] L. Hughes, Roger. (2003). The Flow of Human Crowds. Annu. Rev. Fluid Mech..

35. 169-182. 10.1146/annurev.fluid.35.101101.161136.

[48] Roger L. Hughes, A continuum theory for the flow of pedestrians, Transportation Research Part B: Methodological, Volume 36, Issue 6, 2002.

[49] Erving Goffman. Behavior in public places: Notes on the social organization of

[50] M. Quigley, B. Gerkey, K. Conley, J. Faust, T. Foote, J. Leibs, E. Berger, R. Wheeler, and A. Y. Ng, "ROS: an open-source Robot Operating System," in Proc. Open-Source Software workshop of the International Conference on Robotics and Automation (ICRA), 2009