Abstract
When performing visual servoing or object tracking tasks, active sensor planning is essential to keep targets in sight or to relocate them when missing. In particular, when dealing with a known target missing from the sensor’s field of view, we propose using prior knowledge related to contextual information to estimate its possible location. To this end, this study proposes a Dynamic Bayesian Network that uses contextual information to effectively search for targets. Monte Carlo particle filtering is employed to approximate the posterior probability of the target’s state, from which uncertainty is defined. We define the robot’s utility function via information theoretic formalism as seeking the optimal action which reduces uncertainty of a task, prompting robot agents to investigate the location where the target most likely might exist. Using a context state model, we design the agent’s high-level decision framework using a Partially-Observable Markov Decision Process. Based on the estimated belief state of the context via sequential observations, the robot’s navigation actions are determined to conduct exploratory and detection tasks. By using this multi-modal context model, our agent can effectively handle basic dynamic events, such as obstruction of targets or their absence from the field of view. We implement and demonstrate these capabilities on a mobile robot in real-time.
Similar content being viewed by others
References
Arora P, Papachristos C (2020) Mobile manipulator robot visual servoing and guidance for dynamic target grasping. In: International symposium on visual computing. Springer, pp 223–235
Aydemir A, Pronobis A, Göbelbecker M, Jensfelt P (2013) Active visual object search in unknown environments using uncertain semantics. IEEE Trans Robot 29(4):986–1002
Bai H, Hsu D, Lee W S (2014) Integrated perception and planning in the continuous space: a pomdp approach. Int J Robot Res 33(9):1288–1302
Bajcsy R (1988) Active perception. Proc IEEE 76(8):966–1005
Bertuccelli LF, How JP (2006) Search for dynamic targets with uncertain probability maps. In: 2006 American control conference. IEEE, pp 6–pp
Bourgault F, Furukawa T, Durrant-Whyte HF (2003) Optimal search for a lost target in a bayesian world. In: Field and service robotics. Springer, pp 209–222
Charrow B, Liu S, Kumar V, Michael N (2015) Information-theoretic mapping using cauchy-schwarz quadratic mutual information. In: 2015 IEEE international conference on robotics and automation (ICRA). IEEE, pp 4791–4798
Chung TH, Burdick JW (2012) Analysis of search decision making using probabilistic search strategies. IEEE Trans Robot 28(1):132–144
Denzler J, Brown C M (2002) Information theoretic sensor data selection for active object recognition and state estimation. IEEE Trans Pattern Anal Mach Intell 24(2):145–157
Eidenberger R, Grundmann T, Zoellner R (2009a) Probabilistic action planning for active scene modeling in continuous high-dimensional domains. In: 2009 IEEE International conference on robotics and automation. IEEE, pp 2412–2417
Eidenberger R, Zoellner R, Scharinger J (2009b) An integrated active perception module for a distributed cognitive architecture. In: 2009 International conference on advanced robotics. IEEE, pp 1–7
Freda L, Oriolo G, Vecchioli F (2008) Sensor-based exploration for general robotic systems. In: 2008 IEEE/RSJ International conference on intelligent robots and systems. IEEE, pp 2157– 2164
Hausman K, Kahn G, Patil S, Müller J, Goldberg K, Abbeel P, Sukhatme GS (2016) Occlusion-aware multi-robot 3d tracking. In: 2016 IEEE/RSJ International conference on intelligent robots and systems (IROS). IEEE, pp 1863–1870
Julian B J, Karaman S, Rus D (2014) On mutual information-based control of range sensing robots for mapping applications. Int J Robot Res 33(10):1375–1392
Kaelbling LP, Lozano-Pérez T (2012) Unifying perception, estimation and action for mobile manipulation via belief space planning. In: 2012 IEEE International conference on robotics and automation. IEEE, pp 2952–2959
Kaelbling L P, Lozano-Pérez T (2013) Integrated task and motion planning in belief space. Int J Robot Res 32(9-10):1194– 1227
Krishnamurthy V (2016) Partially observed Markov decision processes. Cambridge University Press
Lau H, Huang S, Dissanayake G (2006) Probabilistic search for a moving target in an indoor environment. In: 2006 IEEE/RSJ International conference on intelligent robots and systems, IEEE, pp 3393–3398
Li JK, Hsu D, Lee WS (2016) Act to see and see to act: Pomdp planning for objects search in clutter. In: 2016 IEEE/RSJ International conference on intelligent robots and systems (IROS). IEEE, pp 5701–5707
Lin TY, Goyal P, Girshick R, He K, Dollár P (2017) Focal loss for dense object detection. In: Proceedings of the IEEE international conference on computer vision, pp 2980–2988
Liu W, Anguelov D, Erhan D, Szegedy C, Reed S, Fu CY, Berg AC (2016) Ssd: Single shot multibox detector. In: European conference on computer vision. Springer, pp 21–37
Luo W, Sun P, Zhong F, Liu W, Zhang T, Wang Y (2019) End-to-end active object tracking and its real-world deployment via reinforcement learning. IEEE Transactions on Pattern Analysis and Machine Intelligence 42(6):1317–1332
Makris A, Kosmopoulos D, Perantonis S, Theodoridis S (2011) A hierarchical feature fusion framework for adaptive visual tracking. Image Vis Comput 29(9):594–606
Porta JM, Vlassis N, Spaan MT, Poupart P (2006) Point-based value iteration for continuous pomdps. J Mach Learn Res 7(Nov):2329–2367
Radmard S, Croft EA (2017) Active target search for high dimensional robotic systems. Auton Robot 41(1):163–180
Radmard S, Meger D, Little JJ, Croft EA (2018) Resolving occlusion in active visual target search of high-dimensional robotic systems. IEEE Trans Robot 34(3):616–629
Redmon J, Farhadi A (2018) Yolov3: An incremental improvement. arXiv:180402767
Redmon J, Divvala S, Girshick R, Farhadi A (2016) You only look once: Unified, real-time object detection. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 779–788
Ross S, Pineau J, Paquet S, Chaib-Draa B (2008) Online planning algorithms for pomdps. J Artif Intell Res 32:663–704
Ryan A, Hedrick JK (2010) Particle filter based information-theoretic active sensing. Robot Auton Syst 58(5):574–584
Shi H, Sun G, Wang Y, Hwang K S (2018) Adaptive image-based visual servoing with temporary loss of the visual signal. IEEE Trans Industr Inform 15(4):1956–1965
Silver D, Veness J (2010) Monte-carlo planning in large pomdps. In: Advances in neural information processing systems, pp 2164–2172
Sridharan M, Wyatt J, Dearden R (2010) Planning to see: a hierarchical approach to planning visual actions on a robot using pomdps. Artif Intell 174(11):704–725
Tseng KS, Mettler B (2017) Near-optimal probabilistic search via submodularity and sparse regression. Auton Robot 41(1):205– 229
Valencia R, Andrade-cetto J (2018) Active pose slam. In: Mapping, planning and exploration with pose SLAM, Springer, pp 89–108
Van Den Berg J, Patil S, Alterovitz R (2012) Motion planning under uncertainty using iterative local optimization in belief space. Int J Robot Res 31(11):1263–1278
Wu K, Ranasinghe R, Dissanayake G (2015) Active recognition and pose estimation of household objects in clutter. In: 2015 IEEE International conference on robotics and automation (ICRA). IEEE, pp 4230–4237
Xiang Y, Alahi A, Savarese S (2015) Learning to track: Online multi-object tracking by decision making. In: Proceedings of the IEEE international conference on computer vision, pp 4705–4713
Ye N, Somani A, Hsu D, Lee WS (2017) Despot: Online pomdp planning with regularization. J Artif Intell Res 58:231–266
Yun S, Choi J, Yoo Y, Yun K, Young Choi J (2017) Action-decision networks for visual tracking with deep reinforcement learning. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 2711–2720
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Electronic supplementary material
Below is the link to the electronic supplementary material.
Appendix: Parameter selection
Appendix: Parameter selection
For particle filtering during experimentation, there are various parameters to be determined, such as the total number of particles, re-sampling conditions (effective sample size) or type of method (residual method), motion model noise, or the sensor model. Here are the parameter list that we empirically determined for our study (Tables 3 and 4).
Rights and permissions
About this article
Cite this article
Kim, M., Sentis, L. Active object tracking using context estimation: handling occlusions and detecting missing targets. Appl Intell 52, 14041–14052 (2022). https://doi.org/10.1007/s10489-021-03116-5
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10489-021-03116-5