Automatic Waypoint Generation to Improve Robot Navigation Through Narrow Spaces
- The identification, without any human intervention, of cumbersome zones in the robot’s working area during an initial inspection stage (typically narrow zones such as passages or doors).
- The automatic and on-the-fly generation of a pair of auxiliary navigation waypoints for each cumbersome zone, which modify the robot trajectory and ensure proper navigation through such zones.
2. Related Work
3. Navigation Assistant
3.1. Automatic Detection of Critical Navigation Points (CNPs)
- The 2D map that was built during the deployment of the robot at home, typically by means of aligning the sensor readings (e.g., from laser scanners) that were recorded during the first exploration of the robot’s operation space.
- The list of nodes or destination points at the different rooms in the house. This is normally performed manually by the operator, selecting those rooms where the robot is expected to carry out some task, while avoiding others like bathrooms where the robot presence may not be welcomed by the end user.
|Algorithm 1: Determining the critical points (CNPs)|
3.2. Generation of Auxiliary Navigation Points (ANPs)
|Algorithm 2: Determining the auxiliary navigation points (ANPs) from the critical points (CNPs).|
4. Experimental Setup
4.1. Laser-Based Maps of Real Houses
- SARMIS: this map corresponds to a (9.8 × 10.7) [m] old-style house with wide rooms and a small corridor. We selected five out of the seven rooms to test navigation, as some of them have a really narrow entrance (up to 54 cm in some cases), which can be potentially not reachable by most robots.
- PARE: this map represents a (10.15 × 9.43) [m] flat-style house with clear square shape and absence of a long, dominant corridor. Discarding one of the bathrooms for the aforementioned reasons, we selected a total of six rooms for testing purposes. The presence of dots in the maps represents a high density of furniture (e.g., tables and chairs) which will indeed make navigation more challenging.
- ANTO: this map has been built from a (12.33 × 8.9) [m] house with a small central corridor that connects the six rooms that compose it. Though challenging, we consider all the rooms of the house for testing navigation.
- MONRY: this map corresponds to a (17.6 × 7.5) [m] elongated shape flat with a dominant, long and narrow corridor where most of the rooms are connected to. It is composed of a total of eight rooms from which we selected seven, discarding again a bathroom.
4.2. Robotics Platforms
- Stevie II : this is the follow-up version of the robot Stevie, who served as a proof of concept of the fact that a socially assistive robot can be deployed in long-term care environments to help seniors and people living with disabilities. Stevie II, which is used within the EPIC EU project, has been built on the project successes and also embodies significant technological upgrades and advanced AI capabilities. It presents a small rectangular footprint (the smallest one in this study), and a height of 130 cm approximately.
- Pepper : developed by Aldebaran Robotics, it is a mobile robot featuring a three-wheeled platform with a triangular footprint. The robot is about 120 cm high, weighs 28 kg and it is equipped with cameras, microphones, speech recognition and social intelligence. It has been designed with the purpose of acting as a companion for the elderly, a teacher of schoolchildren and an assistant in retail shops, among other uses.
- Giraff-X : an active companion robot designed for the assistance of elderly people in their daily life. The Giraff robot, which has been designed and evolved through multiple EU projects: Excite , Giraff-Plus  and MoveCare , is a robotic platform endowed with autonomous navigation capabilities, user interaction, visual object detection, and semantic mapping among its most important skills. Its footprint can be approximated by a circle and contains two motors and two caster wheels, as well as an adjustable height reaching up to 170 cm.
4.3. Navigation Parameters
5. Experimental Results
5.1. Experiment #1: Simulated Robots in Real Maps
- The proposed navigation assistant equals or improves the navigation success rate for all the robot shapes and tested scenarios, reaching improvements of up to 80% (e.g., see Giraff.X results for the SARMIS map, where all the issues were fixed). Nonetheless, when the robot dimensions are too big for a given environment, the improvements are more humble (e.g., see TIAGO results for the SARMIS, PARE or MONRY maps in the last column in Figure 6). These unreachable locations are caused, in most cases, by the reactive planner being unable to calculate a safe navigation path between the ANPs due to the proximity of obstacles, leading to aborting the navigation. Even in these cases, though, our system is capable of increasing the number of successful navigations.
- To ensure an error-free operation of the robot at home, it is recommended to keep a dimension-security-margin of at least 12 cm. That is, the most restrictive robot dimension should be at least 12 cm smaller than the narrowest area in the environment (i.e., doors, corridors, etc.). Both Pepper and Stevie II present little navigation problems given their small footprints in comparison with the house maps used in the experiments (where the narrowest areas corresponded to doors with a size of between 60 cm and 70 cm). Only in a few scenarios, one or two paths are not successfully followed without help (but they are fixed by our assistant), while for the Giraff-X or TIAGO, the navigation failures rise considerably. The reason behind this dimension-security-margin are the multiple sources of error that play a role in the autonomous navigation of the robot, namely: the error related to the laser measurements when sensing the environment, the resolution of the grid-map used to represent the occupancy map (being advisable to increase the resolution as much as possible according to the computational power of the robot), and last but not least, the errors due to the path planning of the robot, where even if a valid global path can be found (i.e., theoretically the robot should be able to pass), its navigation leads to small deviations that can be problematic on too narrow areas.
- When the robot size is close to that of the narrow areas of the environment, a high failure rate in the autonomous navigation is to be expected. It is in these cases where finding the CNPs, either by manually setting them based on expert knowledge or by employing our proposed navigation assistant, becomes mandatory. By forcing the robot to cross those problematic areas in a specific way (i.e., by means of setting the ANPs), high success rates can still be achieved in most scenarios as can be noticed from the results of the Giraff.X robot, where most of the wrong navigations can be solved.
- From the multiple experiments and several navigation attempts, we have learned that the most controversial areas leading to faulty navigation are those involving a narrow area and a circular-like robot trajectory to transverse them. That is, the success rate is usually much higher when the robot is able to plan a path to cross a narrow area (i.e., a doorstep) employing an approximately straight trajectory. See for example the high error rate for Giraff.X or TIAGO when navigating to/from rooms 4, 5 or 6 in the MONRY map, which involves a 90 turning to either enter or leave the room. In contrast, for the same map and robots, room 7 (which has the same doorstep size) does not present problems for the navigation (refer to Figure 7 (bottom), where it can be seen how the entrance to rooms 4, 5 and 6 have been marked as CNPs while room 7 has not).Interestingly, the typical distribution of rooms in a house promotes paths that heavily turn on the doorsteps and corners to reach the different destinations, becoming, therefore, challenging for big sized robots. These scenarios represent the core of our proposal, producing ANPs to enforce the robot to turn in place and traverse the CNPs following paths as straight as possible, hence solving most of the problematic navigations.
- Regarding the automatic detection and characterization of the critical points in the environment, our proposal has demonstrated to be robust and versatile, successfully locating the set of CNPs and ANPs in most situations (see Figure 7). Yet, like any other algorithm, it is not exempt from failures, being advisable the supervision of a technician to ensure maximum coverage of the navigation area within the house during the robot deployment phase, especially with large robots and/or small environments. In any case, our navigation assistant provides a suitable initial proposal of critical zones that, if needed, can be further fine-tuned by a technician or a robotic practitioner.
5.2. Experiment #2: Real Robot in a Real Environment
Conflicts of Interest
- Song, W.K.; Kim, J. Novel assistive robot for self-feeding. In Robotic Systems-Applications, Control and Programming; IntechOpen: Vienna, Austria, 2012; pp. 43–60. [Google Scholar]
- Yanco, H.A. Wheelesley: A robotic wheelchair system: Indoor navigation and user interface. In Assistive Technology and Artificial Intelligence; Springer: Berlin/Heidelberg, Germany, 1998; pp. 256–268. [Google Scholar]
- Brox, E.; Luque, L.F.; Evertsen, G.J.; Hernández, J.E.G. Exergames for elderly: Social exergames to persuade seniors to increase physical activity. In Proceedings of the 2011 5th International Conference on Pervasive Computing Technologies for Healthcare (PervasiveHealth) and Workshops, Dublin, Ireland, 23–26 May 2011; pp. 546–549. [Google Scholar]
- Wüest, S.; Borghese, N.A.; Pirovano, M.; Mainetti, R.; van de Langenberg, R.; de Bruin, E.D. Usability and effects of an exergame-based balance training program. Games Heal. Res. Dev. Clin. Appl. 2014, 3, 106–114. [Google Scholar] [CrossRef] [PubMed][Green Version]
- Ruiz-Sarmiento, J.R.; Galindo, C.; Monroy, J.; Moreno, F.A.; Gonzalez-Jimenez, J. Ontology-based conditional random fields for object recognition. Int. J. Knowledge-Based Syst. 2019, 168, 100–108. [Google Scholar] [CrossRef]
- Monroy, J.; Ruiz-Sarmiento, J.R.; Moreno, F.A.; Melendez-Fernandez, F.; Galindo, C.; Gonzalez-Jimenez, J. A Semantic-Based Gas Source Localization with a Mobile Robot Combining Vision and Chemical Sensing. Sensors 2018, 18, 4174. [Google Scholar] [CrossRef] [PubMed][Green Version]
- Wang, X. Subjective well-being associated with size of social network and social support of elderly. J. Health Psychol. 2016, 21, 1037–1042. [Google Scholar] [CrossRef]
- Orlandini, A.; Kristoffersson, A.; Almquist, L.; Björkman, P.; Cesta, A.; Cortellessa, G.; Galindo, C.; Gonzalez-Jimenez, J.; Gustafsson, K.; Kiselev, A.; et al. ExCITE Project: A Review of Forty-two Months of Robotic Telepresence Technology Evolution. Presence Teleoperators Virtual Environ. 2017. [Google Scholar] [CrossRef][Green Version]
- MoveCare Project. 2019. Available online: http://www.movecare-project.eu/ (accessed on 18 September 2019).
- Antonopoulos, C.; Keramidas, G.; Voros, N.S.; Hübner, M.; Goehringer, D.; Dagioglou, M.; Giannakopoulos, T.; Konstantopoulos, S.; Karkaletsis, V. Robots in assisted living environments as an unobtrusive, efficient, reliable and modular solution for independent ageing: The RADIO perspective. In Proceedings of the International Symposium on Applied Reconfigurable Computing, Bochum, Germany, 13–17 April 2015; pp. 519–530. [Google Scholar]
- Casey, D.; Felzmann, H.; Pegman, G.; Kouroupetroglou, C.; Murphy, K.; Koumpis, A.; Whelan, S. What people with dementia want: Designing MARIO an acceptable robot companion. In Proceedings of the International Conference on Computers Helping People with Special Needs, Linz, Austria, 13–15 July 2016; pp. 318–325. [Google Scholar]
- The Socrates Project. 2019. Available online: http://www.socrates-project.eu/ (accessed on 11 September 2019).
- SECURE—Safety Enables Cooperation in Uncertain Robotic Environments. 2019. Available online: http://secure-robots.eu/ (accessed on 11 September 2019).
- Quigley, M.; Conley, K.; Gerkey, B.; Faust, J.; Foote, T.; Leibs, J.; Wheeler, R.; Ng, A.Y. ROS: An open-source Robot Operating System. In Proceedings of the ICRA Workshop on Open Source Software, Kobe, Japan, 17 May 2009; Volume 3, p. 5. [Google Scholar]
- Lu, D.V.; Hershberger, D.; Smart, W.D. Layered costmaps for context-sensitive navigation. In Proceedings of the 2014 IEEE/RSJ International Conference on Intelligent Robots and Systems, Chicago, IL, USA, 14–18 September 2014; pp. 709–715. [Google Scholar]
- Luperto, M.; Monroy, J.; Ruiz-Sarmiento, J.R.; Moreno, F.A.; Basilico, N.; Gonzalez-Jimenez, J.; Borghese, N.A. Towards Long-Term Deployment of a Mobile Robot for at-Home Ambient Assisted Living of the Elderly. In Proceedings of the European Conference on Mobile Robots, Prague, Czech Republic, 4–6 September 2019. [Google Scholar]
- Kiss, D.; Papp, D. Effective navigation in narrow areas: A planning method for autonomous cars. In Proceedings of the 2017 IEEE 15th International Symposium on Applied Machine Intelligence and Informatics (SAMI), Herl’any, Slovakia, 26–28 January 2017; pp. 000423–000430. [Google Scholar]
- Hsu, D.; Kavraki, L.E.; Latombe, J.C.; Motwani, R.; Sorkin, S. On finding narrow passages with probabilistic roadmap planners. In Robotics: The Algorithmic Perspective: 1998 Workshop on The Algorithmic Foundations of Robotics; A K Peters/CRC Press: New York, NY, USA, 1998; pp. 141–154. [Google Scholar]
- Pan, J.; Manocha, D. Fast probabilistic collision checking for sampling-based motion planning using locality-sensitive hashing. Int. J. Robot. Res. 2016, 35, 1477–1496. [Google Scholar] [CrossRef]
- Pietrzykowski, Z. Ship’s Fuzzy Domain–a Criterion for Navigational Safety in Narrow Fairways. J. Navig. 2008, 61, 499–514. [Google Scholar] [CrossRef]
- Shi, C.; Zhang, M.; Peng, J. Harmonic Potential Field Method for Autonomous Ship Navigation. In Proceedings of the 2007 7th International Conference on ITS Telecommunications, Sophia Antipolis, France, 6–8 June 2007; pp. 1–6. [Google Scholar] [CrossRef]
- Wang, Y.; Chin, H.C. An empirically-calibrated ship domain as a safety criterion for navigation in confined waters. J. Navig. 2016, 69, 257–276. [Google Scholar] [CrossRef][Green Version]
- Fraichard, T. A Short Paper about Motion Safety. In Proceedings of the 2007 IEEE International Conference on Robotics and Automation, Roma, Italy, 10–14 April 2007; pp. 1140–1145. [Google Scholar] [CrossRef]
- Gasparetto, A.; Boscariol, P.; Lanzutti, A.; Vidoni, R. Path planning and trajectory planning algorithms: A general overview. In Motion and Operation Planning of Robotic Systems; Springer: Berlin/Heidelberg, Germany, 2015; pp. 3–27. [Google Scholar]
- Kim, D.; Chung, W.; Park, S. Practical motion planning for car-parking control in narrow environment. IET Control. Theory Appl. 2010, 4, 129–139. [Google Scholar] [CrossRef]
- Banzhaf, H.; Palmieri, L.; Nienhüser, D.; Schamm, T.; Knoop, S.; Zöllner, J.M. Hybrid curvature steer: A novel extend function for sampling-based nonholonomic motion planning in tight environments. In Proceedings of the 2017 IEEE 20th International Conference on Intelligent Transportation Systems (ITSC), Yokohama, Japan, 16–19 October 2017; pp. 1–8. [Google Scholar]
- Wang, W.; Xu, X.; Li, Y.; Song, J.; He, H. Triple RRTs: An Effective Method for Path Planning in Narrow Passages. Adv. Robot. 2010, 24, 943–962. [Google Scholar] [CrossRef][Green Version]
- Hsu, D.; Jiang, T.; Reif, J.; Sun, Z. The bridge test for sampling narrow passages with probabilistic roadmap planners. In Proceedings of the 2003 IEEE International Conference on Robotics and Automation (Cat. No.03CH37422), Taipei, Taiwan, 14–19 September 2003; Volume 3, pp. 4420–4426. [Google Scholar] [CrossRef]
- Lai, T.; Ramos, F.; Francis, G. Balancing Global Exploration and Local-connectivity Exploitation with Rapidly-exploring Random disjointed-Trees. In Proceedings of the 2019 International Conference on Robotics and Automation (ICRA), Montreal, QC, Canada, 20–24 May 2019; pp. 5537–5543. [Google Scholar]
- Borenstein, J.; Wehe, D.; Feng, L.; Koren, Y. Mobile robot navigation in narrow aisles with ultrasonic sensors. In Proceedings of the ANS 6th Topical Meeting on Robotics and Remote Systems, Monterey CA, USA, 5–10 February 1995. [Google Scholar]
- Dai, D.; Jiang, G.; Xin, J.; Gao, X.; Cui, L.; Ou, Y.; Fu, G. Detecting, locating and crossing a door for a wide indoor surveillance robot. In Proceedings of the IEEE International Conference on Robotics and Biomimetics (ROBIO), Shenzhen, China, 12–14 December 2013; pp. 1740–1746. [Google Scholar] [CrossRef]
- Salaris, P.; Vassallo, C.; Souères, P.; Laumond, J.P. The geometry of confocal curves for passing through a door. IEEE Trans. Robot. 2015, 31, 1180–1193. [Google Scholar] [CrossRef]
- Aude, E.P.; Lopes, E.P.; Aguiar, C.S.; Martins, M.F. Door crossing and state identification using robotic vision. IFAC Proc. Vol. 2006, 39, 659–664. [Google Scholar] [CrossRef]
- Cheein, F.A.; De La Cruz, C.; Carelli, R.; Bastos-Filho, T.F. Solution to a door crossing problem for an autonomous wheelchair. In Proceedings of the 2009 IEEE/RSJ International Conference on Intelligent Robots and Systems, St. Louis, MO, USA, 10–15 October 2009; pp. 4931–4936. [Google Scholar]
- Tao, T.; Huang, Y.; Sun, F.; Wang, T. Motion planning for slam based on frontier exploration. In Proceedings of the 2007 International Conference on Mechatronics and Automation, Harbin, China, 5–8 August 2007; pp. 2120–2125. [Google Scholar]
- Kim, B.K.; Tanaka, H.; Sumi, Y. Robotic wheelchair using a high accuracy visual marker lentibar and its application to door crossing navigation. In Proceedings of the 2015 IEEE International Conference on Robotics and Automation (ICRA), Seattle, WA, USA, 26–30 May 2015; pp. 4478–4483. [Google Scholar]
- Tapus, A.; Ramel, G.; Dobler, L.; Siegwart, R. Topology learning and recognition using Bayesian programming for mobile robot navigation. In Proceedings of the 2004 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) (IEEE Cat. No. 04CH37566), Sendai, Japan, 28 September–2 October 2004; Volume 4, pp. 3139–3144. [Google Scholar]
- Althaus, P.; Christensen, H.I. Smooth task switching through behaviour competition. Robot. Auton. Syst. 2003, 44, 241–249. [Google Scholar] [CrossRef]
- Ravankar, A.; Ravankar, A.; Kobayashi, Y.; Emaru, T. Intelligent Robot Guidance in Fixed External Camera Network for Navigation in Crowded and Narrow Passages. Proceedings 2016, 1, 37. [Google Scholar] [CrossRef][Green Version]
- García-Soler, Á.; Facal, D.; Díaz-Orueta, U.; Pigini, L.; Blasi, L.; Qiu, R. Inclusion of service robots in the daily lives of frail older users: A step-by-step definition procedure on users’ requirements. Arch. Gerontol. Geriatr. 2018, 74, 191–196. [Google Scholar] [CrossRef]
- Mandow, A.; Gomez-de-Gabriel, J.M.; Martinez, J.L.; Munoz, V.F.; Ollero, A.; Garcia-Cerezo, A. The autonomous mobile robot AURORA for greenhouse operation. IEEE Robot. Autom. Mag. 1996, 3, 18–28. [Google Scholar] [CrossRef][Green Version]
- Kuo, C.H.; Chen, H.H. Human-Oriented Design of Autonomous Navigation Assisted Robotic Wheelchair for Indoor Environments. In Proceedings of the 2006 IEEE International Conference on Mechatronics, Budapest, Hungary, 3–5 July 2006; pp. 230–235. [Google Scholar] [CrossRef]
- Savkin, A.V.; Wang, C. A framework for safe assisted navigation of semi-autonomous vehicles among moving and steady obstacles. Robotica 2017, 35, 981–1005. [Google Scholar] [CrossRef]
- Jaillet, L.; Cortés, J.; Siméon, T. Sampling-based path planning on configuration-space costmaps. IEEE Trans. Robot. 2010, 26, 635–646. [Google Scholar] [CrossRef][Green Version]
- Mainprice, J.; Sisbot, E.A.; Jaillet, L.; Cortés, J.; Alami, R.; Siméon, T. Planning human-aware motions using a sampling-based costmap planner. In Proceedings of the 2011 IEEE International Conference on Robotics and Automation, Shanghai, China, 9–13 May 2011; pp. 5012–5017. [Google Scholar]
- Grisetti, G.; Stachniss, C.; Burgard, W. Improved techniques for grid mapping with rao-blackwellized particle filters. IEEE Trans. Robot. 2007, 23, 34. [Google Scholar] [CrossRef][Green Version]
- Jaimez, M.; Monroy, J.; Lopez-Antequera, M.; Gonzalez-Jimenez, J. Robust Planar Odometry based on Symmetric Range Flow and Multi-Scan Alignment. IEEE Trans. Robot. 2018, 1623–1635. [Google Scholar] [CrossRef]
- Ruiz-Sarmiento, J.R.; Galindo, C.; González-Jiménez, J. Robot@Home, a Robotic Dataset for Semantic Mapping of Home Environments. Int. J. Robot. Res. 2017, 36, 131–141. [Google Scholar] [CrossRef][Green Version]
- McGinn, C.; Bourke, E.; Murtagh, A.; Cullinan, M.; Kelly, K. Exploring the application of design thinking to the development of service robot technology. In Proceedings of the ICRA2018 Workshop on Elderly Care Robotics-Technology and Ethics (WELCARO), Brisbane, Australia, 20–25 May 2018. [Google Scholar]
- Meet Pepper the Emotional Robot. 2014. Available online: http://edition.cnn.com/2014/06/06/tech/innovation/pepper-robot-emotions/ (accessed on 11 September 2019).
- EXCITE Project. 2012. Available online: http://www.aal-europe.eu/projects/excite/ (accessed on 18 September 2019).
- GiraffPlus Project. 2015. Available online: http://www.giraffplus.eu/ (accessed on 18 September 2019).
- Pages, J.; Marchionni, L.; Ferro, F. Tiago: The modular robot that adapts to different research needs. In Proceedings of the International Workshop on Robot Modularity, IROS, Daejeon, Korea, 9–14 October 2016. [Google Scholar]
- ENRICHME Project. 2018. Available online: https://cordis.europa.eu/project/rcn/194090/factsheet/en (accessed on 18 September 2019).
- GrowMeUp Project. 2018. Available online: https://cordis.europa.eu/project/rcn/194088/factsheet/en (accessed on 18 September 2019).
- Guimarães, R.L.; de Oliveira, A.S.; Fabro, J.A.; Becker, T.; Brenner, V.A. ROS navigation: Concepts and tutorial. In Robot Operating System (ROS); Springer: Berlin/Heidelberg, Germany, 2016; pp. 121–160. [Google Scholar]
- Gerkey, B.P. AMCL Reference Website. 2019. Available online: http://wiki.ros.org/amcl (accessed on 26 November 2019).
- Fox, D.; Burgard, W.; Dellaert, F.; Thrun, S. Monte carlo localization: Efficient position estimation for mobile robots. In Proceedings of the Sixteenth National Conference on Artificial Intelligence (AAAI ’99), Orlando, FL, USA, 18–22 July 1999. [Google Scholar]
- Marder-Eppstein, E. Move_Base Reference Website. 2019. Available online: http://wiki.ros.org/move_base (accessed on 26 November 2019).
- Vaughan, R. Massively multi-robot simulation in stage. Swarm Intell. 2008, 2, 189–208. [Google Scholar] [CrossRef]
© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
Moreno, F.-A.; Monroy, J.; Ruiz-Sarmiento, J.-R.; Galindo, C.; Gonzalez-Jimenez, J. Automatic Waypoint Generation to Improve Robot Navigation Through Narrow Spaces. Sensors 2020, 20, 240. https://doi.org/10.3390/s20010240
Moreno F-A, Monroy J, Ruiz-Sarmiento J-R, Galindo C, Gonzalez-Jimenez J. Automatic Waypoint Generation to Improve Robot Navigation Through Narrow Spaces. Sensors. 2020; 20(1):240. https://doi.org/10.3390/s20010240Chicago/Turabian Style
Moreno, Francisco-Angel, Javier Monroy, Jose-Raul Ruiz-Sarmiento, Cipriano Galindo, and Javier Gonzalez-Jimenez. 2020. "Automatic Waypoint Generation to Improve Robot Navigation Through Narrow Spaces" Sensors 20, no. 1: 240. https://doi.org/10.3390/s20010240