Initial localization by set inversion

Question

1. What are the contributions mentioned in the paper "Initial localization by set inversion" ?

2. What is the significance of the approach?

3. What is the way to represent the environment?

4. What are the only assumptions that are required to solve the initial localization problem?

Accepted Answer

In this paper, initial localization problems are solved by using set-membership estimation.. The method can be used with any robot and any kind of sensor ( s ), provided that a computable model of the environment/sensor interaction is available.. Salient properties of the method are as follows.

Accepted Answer

A remarkable feature of the approach lies in the fact that, when measurements are ambiguous because of symmetries in the environment, the estimate of the configuration domain may be disconnected, each connected component corresponding to a specific matching hypothesis.

Accepted Answer

When the task is clearly defined, using analytical expressions to express the relations between the mobile and the items of the environment saves memory space and yields localization methods that are potentially less time consuming than cell-based representations [4].

Accepted Answer

The only assumptions are the availability of 1) a sufficiently representative map, and 2) a computable model of the interaction between the sensors and the environment (both constitute an observation simulator).

Accepted Answer

A deterministic observation simulator is a function s (12) computing the measurements that should be reported by the n sensors if the configuration of the vehicle were q, assuming that the models of the sensors and of the environment are correct.

Accepted Answer

The problem to be solved is the guaranteed estimation of the reciprocal image Q of a feasible measurement set rI (13) by the simulator function s described by (12).

Accepted Answer

Drumheller [16] introduced the concept of an interpretation tree in the context of localization, and designed a strategy to prune this interpretation tree to limit complexity.

Accepted Answer

The computational complexity of the algorithm is bilinear in the number of sensors and in the number of environment items that constitute the map; however, it is exponential in the dimension of the configuration to be estimated.

Accepted Answer

When there are several possible initial estimates, multitarget tracking [8] can be used to extend the original localization mechanism [9].

Accepted Answer

For instance, analytical representations are most frequently used in sensor-referenced control where the task to be performed is implemented by output feedback regulation algorithms.

Accepted Answer

In [21], this Markovian approach is used, together with an optimization procedure that estimates the most probable cell inside the updated grid; superimposing interpolation functions allows the precision of the estimate to be increased.

Accepted Answer

The practical computation of the intersection [ni=1S (di; 0 i) of the subsets S (di; 0i) corresponding to n different measurements obtained at the same pose is the principle of the method developed in the sequeld 2 [d d; d+ d]2 [ 0 ; 0 + ]l2 ! MP ; !

Accepted Answer

One possibility for allowing the presence of outliers in the measurements is to relax the conditions for a box to be feasible when using SIVIA (Section III-B).

Accepted Answer

As all configurations in the solution set are equally candidates to be the true one, the strategy to move the robot under this knowledge should minimize a risk criterion rather than maximize an observability criterion.

Accepted Answer

Using fuzzy sets in place of usual sets, in the direction proposed by [29], would be an alternative strategy to take data uncertainty into account.

Initial localization by set inversion

Chat with Paper

AI Agents for this Paper

Most frequently asked questions

1. What are the contributions mentioned in the paper "Initial localization by set inversion" ?

2. What is the significance of the approach?

3. What is the way to represent the environment?

4. What are the only assumptions that are required to solve the initial localization problem?

5. What is the definition of a deterministic observation simulator?

6. What is the problem to be solved by the simulator function s?

7. How did Drumheller design the interpretation tree?

8. What is the computational complexity of the algorithm?

9. What is the method for extending the initial estimate?

10. What is the common use of analytical representations in the environment?

11. What is the common method used to estimate the location of a robot?

12. What is the principle of the deterministic observation simulator?

13. What is the way to allow outliers to be observed?

14. What is the way to move a robot?

15. What is the alternative strategy to take data uncertainty into account?

Figures

Citations

Brief paper: Robust set-membership state estimation; application to underwater robotics

A Nonlinear Set Membership Approach for the Localization and Map Building of Underwater Robots

Localization of a mobile robot using relative bearing measurements

Constraints propagation techniques on intervals for a guaranteed localization using redundant data

Passive mobile robot localization within a fixed beacon field

References

Tracking and data association

Methods and applications of interval analysis

Tracking and data association

Using occupancy grids for mobile robot perception and navigation

Applied Interval Analysis

Related Papers (5)

Applied Interval Analysis

Probabilistic Robotics

Methods and applications of interval analysis

Brief paper: Robust set-membership state estimation; application to underwater robotics

Contractor programming