Searcharxiv⌕ Search
Home/Archive/arXiv

arXiv · 2111.12579

Water Care: Water Surface Cleaning Bot and Water Body Surveillance System

Harsh Sankar Naicker·Yash Srivastava·Akshara Pramod·Niket Paresh Ganatra·Deepakshi Sood·Saumya Singh·Velmathi Guruviah
View original source ↗Read source document ↗Export metadata ↓

Abstract

Whenever a person hears about pollution, more often than not, the first thought that comes to their mind is air pollution. One of the most under-mentioned and under-discussed pollution globally is that caused by the non-biodegradable waste in our water bodies. In the case of India, there is a lot of plastic waste on the surface of rivers and lakes. The Ganga river is one of the 10 rivers which account for 90 percent of the plastic that ends up in the sea and there are major cases of local nalaas and lakes being contaminated due to this waste. This limits the source of clean water which leads to major depletion in water sources. From 2001 to 2012, in the city of Hyderabad, 3245 hectares of lakes dissipated. The water recedes by nine feet a year on average in southern New Delhi. Thus, cleaning of these local water bodies and rivers is of utmost importance. Our aim is to develop a water surface cleaning bot that is deployed across the shore. The bot will detect garbage patches on its way and collect the garbage thus making the water bodies clean. This solution employs a surveillance mechanism in order to alert the authorities in case anyone is found polluting the water bodies. A more sustainable system by using solar energy to power the system has been developed. Computer vision algorithms are used for detecting trash on the surface of the water. This trash is collected by the bot and is disposed of at a designated location. In addition to cleaning the water bodies, preventive measures have been also implemented with the help of a virtual fencing algorithm that alerts the authorities if anyone tries to pollute the water premises. A web application and a mobile app is deployed to keep a check on the movement of the bot and shore surveillance respectively. This complete solution involves both preventive and curative measures that are required for water care.

Explore related subjects

cs.ROcs.CV

Keep this discovery

BibTeXRIS

Harsh Sankar Naicker, Yash Srivastava, Akshara Pramod, Niket Paresh Ganatra, Deepakshi Sood, Saumya Singh, Velmathi Guruviah. 2021-11-24. Water Care: Water Surface Cleaning Bot and Water Body Surveillance System. https://arxiv.org/abs/2111.12579

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

IMLE-VLA: Fast Single-Step Action Generation for Vision-Language-Action Policies

Vision-language-action (VLA) policies leverage pretrained vision-language backbones to achieve strong cross-task generalization. A leading design couples this backbone with a dedicated continuous action head trained via diffusion or flow matching. However, such heads rely on iterative multi-step sampling, for example 10 Euler steps in $\pi_{0.5}$. This creates an inference bottleneck that produces stop-and-go movement in the robot and slower task completion. We introduce IMLE-VLA, which replaces the iterative action head with a single-step conditional generator trained via conditional Implicit Maximum Likelihood Estimation (cIMLE). The cIMLE objective promotes multimodal action coverage, avoiding the mode collapse of naive regression heads while eliminating multi-step sampling entirely. When IMLE-VLA is applied to $\pi_{0.5}$, it increases inference frequency 3.67x (55 Hz vs. 15 Hz), enabling up to 11x higher action throughput. On the 40-task LIBERO benchmark, IMLE-VLA achieves the highest average success rate (98.0%) among all baselines while leading in inference frequency. Under the test-time perturbations of LIBERO-plus, IMLE-VLA retains $\pi_{0.5}$'s robustness while other baselines degrade sharply, confirming that the cIMLE head preserves generalization. Real-world experiments on a Franka Emika Panda across four tasks demonstrate smoother motion (2.2x to 3.0x lower jerk) and faster task completion, with IMLE-VLA outperforming $\pi_{0.5}$ on every task and reducing average VLA inference time per episode by 3.9x to 6.6x. Videos and code are available at https://kianhk6.github.io/IMLE-VLA/

cs.RO↗
arXiv2026

ObstaDiff: Generalizable Diffusion Policy Learning via Obstacle-aware Representations

Imitation learning has achieved impressive results in robotic manipulation, yet most existing approaches assume clean backgrounds and lack explicit mechanisms for obstacle-aware motion generation. Extending such policies to cluttered, real-world scenes with unstructured obstacles remains a key generalization challenge. We present ObstaDiff, a decomposed diffusion-policy framework with a lightweight obstacle-aware visual encoder. ObstaDiff extracts a structured target-obstacle-background representation, enabling the downstream alignment policy to generate end-effector trajectories toward a target-centered bottleneck pose while reasoning about surrounding obstacles. We evaluate ObstaDiff on 61 real-robot greenhouse trials per method (366 executions in total). ObstaDiff achieves 75.41% average task success and 8.20% average obstacle collision rate, outperforming representative imitation-learning baselines and improving generalization in cluttered agricultural scenes.

cs.RO↗
arXiv2026

Testing Between the Test Cases: Proving End-to-End Steering in Conditions You Never Drove

AI-based automated vehicle testing is challenging because a model that passes every test condition can still fail in the real world. Formal verification offers a way to directly address this gap. On a simulated highway and an arterial road we trained two small end-to-end steering networks each in CARLA, one on clear conditions alone and one on clear, fog, night and low sun. All four models were driven against a 2.19 ft lane-departure budget. Without driving again, we used bound propagation, a formal method that reads the trained weights, to compute how far steering can drift at every disturbance strength between two captured images. One calculation covers more than a campaign could drive: on the arterial it spans 133 poses, where ten intensities each would be 10^133 combinations, in minutes on one GPU. Not only did formal verification find conditions that broke the clear-trained policy without simulation testing, it provided some preliminary evidence for potential failures between the test cases. Our overall conclusion is that formal verification is a viable complement to simulation, and could be adopted as a part of verification and validation for automated driving.

cs.RO↗