# Non-Linear Estimation using the Weighted Average Consensus-Based Unscented Filtering for Various Vehicles Dynamics towards Autonomous Sensorless Design

**Authors:** Bambang L. Widjiantoro, Moh Kamalul Wafi, Katherin Indriawati  
**arXiv:** [2304.04766](https://arxiv.org/abs/2304.04766)  
**Date:** 2023-04-09  
**Summary page:** https://www.summarizepaper.com/en/paper/2304.04766v1/

## Key points

- Autonomous vehicles raise concerns about coping with environmentally dynamic non-linear systems under constraints and disturbances
- These vehicles are connected to their own instruments and components in their neighborhoods, creating diverse interconnected communications that need to be handled locally for faster decision-making
- Distributed estimation is used to reach untouched states, pursuing sensorless design
- Modified pseudo measurement is constructed to lead to weighted average consensus calculation within unscented filtering along with bounded estimation errors
- Proposed algorithm is tested on various robust control scenarios subject to noise and disturbance, with stability analysis ensuring its effectiveness
- Numerical instances are presented along with the performances of the control and estimation method, which affirms its effectiveness with limited error deviation compared to other centralized and distributed filtering methods
- Study weighs the performances of the estimation according to reviews of distributed estimation and various applications in terms of non-linear uncertainties interconnected system and Pareto optimization
- Results from arbitrary initial conditions highlight convergence almost identical as the reference (r)
- Further research is suggested for directed sensorless design and fault-tolerant learning control subject to faults that negate failures.
- Overall, this study presents a novel approach for non-linear estimation using weighted average consensus-based unscented filtering for various vehicle dynamics towards autonomous sensorless design. The results demonstrate its effectiveness in dealing with interconnected networks while providing insights into future research directions for improving fault tolerance in autonomous vehicles.

## Summary

The development of autonomous vehicles has raised concerns about coping with the environmentally dynamic non-linear systems under constraints and disturbances. These vehicles are not only connected to their own instruments but also to the components in their neighborhoods, creating diverse interconnected communications that need to be handled locally to ease computation and fasten decision-making. To address these challenges, distributed estimation is used to reach untouched states, pursuing sensorless design. This approach involves constructing a modified pseudo measurement that leads to weighted average consensus calculation within unscented filtering along with bounded estimation errors. The proposed algorithm is tested on various robust control scenarios subject to noise and disturbance, with stability analysis ensuring its effectiveness. Numerical instances are presented along with the performances of the control and estimation method, which affirms its effectiveness with limited error deviation compared to other centralized and distributed filtering methods. In addition, this study weighs the performances of the estimation according to reviews of distributed estimation and various applications in terms of non-linear uncertainties interconnected system and Pareto optimization. The results from arbitrary initial conditions highlight convergence almost identical as the reference (r). Moreover, further research is suggested for directed sensorless design and fault-tolerant learning control subject to faults that negate failures. Overall, this study presents a novel approach for non-linear estimation using weighted average consensus-based unscented filtering for various vehicle dynamics towards autonomous sensorless design. The results demonstrate its effectiveness in dealing with interconnected networks while providing insights into future research directions for improving fault tolerance in autonomous vehicles.