Indoor Navigation using Consumer Portable Devices in Cart/Stroller
- Sep 25, 2017
- 2 min read
CASE STUDY
In the past few years, the huge and rapid development of portable devices (e.g. smartphones) as well as the increase in the number of users motivated researchers to effectively use the various components of such devices to develop different applications.
One emerging application using smartphones is to navigate users in shopping centers or retail stores while they use a shopping cart or stroller. The smartphone could be placed in the cart/stroller, whether directly, or inside a bag or purse in the cart/stroller or attached/hanged to the cart/stroller while it is pushed or pulled.
Cart Navigation using portable devices is a challenging task since INS will not provide acceptable accuracy, and the conventional PDR is not appropriate to be applied to such case as the device is not on the user body or directly related to it. The required information can’t be obtained or extracted from the collected data without a pre-processing step using appropriate signal processing techniques and disturbance/noise filtering.
In this paper, an enhanced dead reckoning positioning technique for cart navigation using portable devices is proposed. The proposed technique for cart navigation relies on dead reckoning and uses data from accelerometers and gyroscopes sensors. Also, any available absolute positioning information, upon availability, can be integrated with the proposed cart dead reckoning technique for further enhancement to the navigation solution.
Various tests scenarios were conducted to assess the performance of the proposed cart dead reckoning technique. The datasets were collected by a variety of subjects using different types of carts and strollers with various smartphones in different orientations and misalignment angles at different speeds of users.
The obtained results for over 1000 cart/stroller data demonstrate the capabilities of the proposed technique and its suitability for retail stores indoor navigation using portable devices in cart/stroller. The proposed algorithm shows competitive positioning performance of the portable devices in cart in terms of accuracy, reliability, and availability.
Read the full paper here.
I keep coming back to Pixel Flow because every level feels readable, even when it’s hard. You’ve got the layered block object right in front of you, the conveyor queue coming in, the colors of the pigs, and the waiting slots counting down your room for error. There’s enough going on to make it tense, but it never feels like the game is hiding information. That matters to me.
I finally used the Move Box booster in beads out on a blocked setup and it felt like cheating, but also like exactly the kind of bailout the board was begging for.