On the Accuracy and Efficiency of Sensing and Localization for Robotics
- 16 November 2020
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 21 (7) , 2480-2492
Abstract
In recent robotic applications, a critical need is to simultaneously detect communication (emission state) and estimate its trajectory. Whilst wireless sensor observations are useful, they are often uncertain due to the stochastic communication bursts and robot mobility. Over-sampling the information environment can incur excessive radio interference and energy usage. Therefore, one challenge is how to improve the efficiency of sensing under sparse and dynamic information, and make accurate inference on the robot's location. Here, we design a novel mixed detection and estimation (MDE) scheme to enhance both the accuracy and the efficiency by exploiting the mobility pattern correlations. Relying on a Markov state-space model, dynamic behaviors of robot's emission state and movement are formulated. A two-stage sequential Bayesian scheme, premised on random finite set (RFS), is developed to detect and estimate the involved unknown states. Specifically, in order to counteract the probability likelihood disappearance (caused by no information emission) and improve robustness to ambient noise, a sequential pre-filtering technique is designed, which can refine local observations and thereby significantly improve the accuracy of the system. We validate the proposed MDE scheme via both theoretical analysis and numerical simulations, demonstrating it would improve both the detection and estimation accuracy and efficiency.Keywords
Funding Information
- National Natural Science Foundation of China (61471061)
- National Natural Science Foundation of China (U1805262)
- International Exchanges Scheme of NSFC
- Royal Society (6151101238)
- Royal Society (IE150708)
This publication has 33 references indexed in Scilit:
- Formulation and Analysis of LMS Adaptive Networks for Distributed Estimation in the Presence of Transmission ErrorsIEEE Internet of Things Journal, 2015
- Energy Detection Based Spectrum Sensing for Cognitive Radios Over Time-Frequency Doubly Selective Fading ChannelsIEEE Transactions on Signal Processing, 2014
- Maximizing the value of sensed information in underwater wireless sensor networks via an autonomous underwater vehicle2014
- A distributed energy-efficient clustering algorithm with improved coverage in wireless sensor networksFuture Generation Computer Systems, 2012
- Gaussian Approximations for Energy-Based Detection and Localization in Sensor Networks2007
- Particle FilteringIEEE Signal Processing Magazine, 2003
- Relative location estimation in wireless sensor networksIEEE Transactions on Signal Processing, 2003
- An application-specific protocol architecture for wireless microsensor networksIEEE Transactions on Wireless Communications, 2002
- A survey of mobility models for ad hoc network researchWireless Communications and Mobile Computing, 2002
- Comparison of Kalman and least squares filters for locating autonomous very low frequency acoustic sensorsIEEE Journal of Oceanic Engineering, 1988