Akhil Jayadeep

Robot-aided outdoor cleaning is a challenging task with plenty of room for improvement. The cleaning profile of a robot should be adjusted based on the availability of litter to ensure proper and efficient cleaning. This article proposes a novel system that integrates a Glasius Bio-Inspired Neural Network (GBNN) for coverage path planning with a vision-based cleaning profile adaptation scheme. The vision-based adaptation occurs based on the presence of leaves on the ground. The proposed system operates in two phases: an initial phase with low-power cleaning to cover areas with no leaves, while localizing leaf spots that require high-power cleaning, and a secondary phase focusing on cleaning these detected leaf spots using a high-power cleaning profile. This cleaning profile adaptation enables more efficient and effective cleaning. Simulation results showed a 47% improvement in energy efficiency for the proposed method compared to a method with no cleaning profile adaptation. Robot hardware tests conducted using the Panthera 2.0 outdoor cleaning robot have demonstrated the real-world applicability of the proposed method.

The demand for large-scale, heavy-duty autonomous pavement-sweeping robots is rising due to urban growth, hygiene needs, and labor shortages. Ensuring their health and safe operation in dynamic outdoor environments is vital, as terrain unevenness and slope gradients can accelerate wear, increase maintenance costs, and pose safety risks. This study introduces an AI-driven condition monitoring (CM) framework designed to detect terrain unevenness and slope gradients in real time, distinguishing between safe and unsafe conditions. As system vibration levels and energy consumption vary with terrain unevenness and slope gradients, vibration and current data are collected for five CM classes identified: safe, moderately safe terrain, moderately safe slope, unsafe terrain, and unsafe slope. A simple-structured one-dimensional convolutional neural network (1D CNN) model is developed for fast and accurate prediction of the safe to unsafe classes for real-time application. An in-house developed large-scale autonomous pavement-sweeping robot, PANTHERA 2.0, is used for data collection and real-time experiments. The training dataset is generated by extracting representative vibration and heterogeneous slope data using three types of interoceptive sensors mounted in different zones of the robot. These sensors complement each other to enable accurate class prediction. The dataset includes angular velocity data from an IMU, vibration acceleration data from three vibration sensors, and current consumption data from three current sensors attached to the key motors. A CM-map framework is developed for real-time monitoring of the robot by fusing the predicted anomalous classes onto a 3D occupancy map of the workspace. The performance of the trained CM framework is evaluated through offline and real-time field trials using statistical measurement metrics, achieving an average class prediction accuracy of 92% and 90.8%, respectively. This demonstrates that the proposed CM framework enables maintenance teams to take timely and appropriate actions, including the adoption of suitable maintenance strategies.

Outdoor cleaning robots must operate reliably across diverse and unstructured surfaces, yet many existing systems lack the adaptability to handle terrain variability. This paper proposes a terrain-aware cleaning framework that dynamically adjusts robot behavior based on real-time surface classification and slope estimation. A 128-channel LiDAR sensor captures signal intensity images, which are processed by a ResNet-18 convolutional neural network to classify floor types as wood, smooth, or rough. Simultaneously, pitch angles from an onboard IMU detect terrain inclination. These inputs are transformed into fuzzy sets and evaluated using a Mamdani-type fuzzy inference system. The controller adjusts brush height, brush speed, and robot velocity through 81 rules derived from 48 structured cleaning experiments across varying terrain and slopes. Validation was conducted in low-light (night-time) conditions, leveraging LiDAR’s lighting-invariant capabilities. Field trials confirm that the robot responds effectively to environmental conditions, such as reducing speed on slopes or increasing brush pressure on rough surfaces. The integration of deep learning and fuzzy control enables safe, energy-efficient, and adaptive cleaning in complex outdoor environments. This work demonstrates the feasibility and real-world applicability for combining perception and inference-based control in terrain-adaptive robotic systems.

Outdoor sweeping is a tedious and labor-intensive task essential for maintaining the cleanliness of public spaces such as gardens and parks. Robots have been developed to address the limitations of traditional methods. Coverage Path Planning (CPP) is a critical function for these robots. However, existing CPP methods often perform poorly in large environments, where such robots are typically deployed. This paper proposes a novel CPP framework for outdoor sweeping robots operating in expansive outdoor areas, defined as environments exceeding 1000 square meters in size. The framework begins by decomposing the environment into smaller sub-regions. The sequence in which these sub-regions are visited is then optimized by formulating the problem as a Travelling Salesman Problem (TSP), aiming to minimize travel distance. Once the visiting sequence is determined, a boustrophedon-based CPP is applied within each sub-region. We analyzed two decomposition strategies, Voronoi-based and grid-based, and evaluated three TSP optimization techniques: local search, record-to-record travel, and simulated annealing. This results in six possible combinations. Simulation results demonstrated that Voronoi-based decomposition achieves higher area coverage (average coverage of 95.6%) than grid-based decomposition (average coverage 52.8%). For Voronoi-based methods, local search yielded the shortest computation time, while simulated annealing achieved the lowest travel distance. We have also conducted hardware experiments to validate the real-world applicability of the proposed framework for efficient CPP in outdoor sweeping robots. The robot hardware experiment achieved 84% coverage in a 19 m × 17 m environment.