Path planning and navigation is a very important problem in robotics, especially for mobile robots operating in complex environments. Sampling based planners such as the probabilistic roadmaps (PRM) have been widely used for different robot applications. However, due to the random sampling of nodes in PRM, it suffers from narrow passage problem that generates unconnected graph. The problem is addressed by increasing the number of nodes but at higher computation cost affecting real-time performance. To address this issue, in this paper, we propose an improved sampling-based path planning method for mobile robot navigation. The proposed method uses a layered hybrid Probabilistic Roadmap (PRM) and the Artificial Potential Field (APF) method for global planning. We used a decomposition method for node distribution that uses map segmentation to produce regions of high and low potential, and propose a method of reducing the dispersion of sample set during the roadmap construction. Our method produces better goal planning queries with a smaller graph and is computationally efficient than the traditional PRM. The proposed planner called the Hybrid Potential based Probabilistic Roadmap (HPPRM) is an improved sampling method with respect to success rate and calculation cost. Furthermore, we present a method for reactive local motion planning in the presence of static and dynamic obstacles in the environment. The advantage of the proposed method is that it can avoid local minima and successfully generate plans in complex maps such as narrow passages and bug trap scenarios that are otherwise difficult for the traditional sample-based methods. We show the validity of our method with experiments in simulation and real environments for both local and global planning. The results indicate that the proposed HPPRM is effective for autonomous mobile robot navigation in complex environments. The success rate of the proposed method is higher than 95% both for local and global planning.

In recent years, autonomous vehicle driving technology and advanced driver assistance systems have played a key role in improving road safety. However, weather conditions such as snow pose severe challenges for autonomous driving and are an active research area. Thanks to their superior reliability, the resilience of detection, and improved accuracy, advances in computation and sensor technology have paved the way for deep learning and neural network–based techniques that can replace the classical approaches. In this research, we investigate the semantic segmentation of roads in snowy environments. We propose a multi-modal fused RGB-T semantic segmentation utilizing a color (RGB) image and thermal map (T) as inputs for the network. This paper introduces a novel fusion module that combines the feature map from both inputs. We evaluate the proposed model on a new snow dataset that we collected and on other publicly available datasets. The segmentation results show that the proposed fused RGB-T input can segregate human subjects in snowy environments better than an RGB-only input. The fusion module plays a vital role in improving the efficiency of multiple input neural networks for person detection. Our results show that the proposed network can generate a higher success rate than other state-of-the-art networks. The combination of our fused module and pyramid supervision path generated the best results in both mean accuracy and mean intersection over union in every dataset.

Heterogeneous robot teams offer significant advantages for exploration and navigation missions in unknown environments. Notably, the air-ground robot teams have superior efficiency due to their mixed abilities and shared sensor resources useful when completing missions. However, the high energy consumption and reduced payload capability of the vertical take-off and landing (VTOL) aerial platforms limit the complete robot team's behavior. At some point in a mission, the aircraft needs to return to the base, land, and replenish its batteries. This task restricts the team to explore the environment in-depth and to work for longer durations. To solve this problem, we present an autonomous system for docking a VTOL-unmanned aerial vehicles (UAVs) with a mobile manipulator. The central measurement-actuation unit consists of a robot manipulator mounted on a mobile platform with a visual sensor configured as an eye-in-hand device. The visual information is used to execute a stable UAV tracking to achieve air-ground robot contact firmly. This strategy allows the multirobot team to economize and potentially recover aircraft energy. It also enables the robust ground platform to protect and store the aircraft facilitating to continue the mission for a longer duration. The on-site autonomous docking can be initiated multiple times, allowing the team to explore larger areas and complete long-term missions. The proposed method has been tested in simulations and real environments with detailed experimental results.

Large scale operational areas often require multiple service robots for coverage and task parallelism. In such scenarios, each robot keeps its individual map of the environment and serves specific areas of the map at different times. We propose a knowledge sharing mechanism for multiple robots in which one robot can inform other robots about the changes in map, like path blockage, or new static obstacles, encountered at specific areas of the map. This symbiotic information sharing allows the robots to update remote areas of the map without having to explicitly navigate those areas, and plan efficient paths. A node representation of paths is presented for seamless sharing of blocked path information. The transience of obstacles is modeled to track obstacles which might have been removed. A lazy information update scheme is presented in which only relevant information affecting the current task is updated for efficiency. The advantages of the proposed method for path planning are discussed against traditional method with experimental results in both simulation and real environments.

Hitchhiking is a means of transportation gained by asking other people for a (free) ride. We developed a multi-robot system which is the first of its kind to incorporate hitchhiking in robotics, and discuss its advantages. Our method allows the hitchhiker robot to skip redundant computations in navigation like path planning, localization, obstacle avoidance, and map update by completely relying on the driver robot. This allows the hitchhiker robot, which performs only visual servoing, to save computation while navigating on the common path with the driver robot. The driver robot, in the proposed system performs all the heavy computations in navigation and updates the hitchhiker about the current localized positions and new obstacle positions in the map. The proposed system is robust to recover from `driver-lost’ scenario which occurs due to visual servoing failure. We demonstrate robot hitchhiking in real environments considering factors like service-time and task priority with different start and goal configurations of the driver and hitchhiker robots. We also discuss the admissible characteristics of the hitchhiker, when hitchhiking should be allowed and when not, through experimental results.

This paper presents a novel bio-inspired hybrid communication framework that incorporates the repelling behaviour of anti-aphrodisiac pheromones and attractive behaviour of pheromones for efficient map exploration of multiple mobile service robots. The proposed communication framework presents a scheme for robots to efficiently serve large areas of map, while cooperating with each other through proper pheromone deposition. This eliminates the need of explicitly programming each service robot to serve particular areas of the map. The paths taken by robots are represented as nodes across which pheromones are deposited. This reduces the search space for tracking pheromones and reduces data size to be communicated between robots. A novel pheromone deposition model is presented which takes into account the uncertainty in the robot’s position. This eliminates robots to deposit pheromones at wrong places when localization fails. The framework also integrates the pheromone signalling mechanism in landmark-based Extended Kalman Filter (EKF) localization and allows the robots to capture areas or sub-areas of the map, to improve the localization. A scheme to resolve conflicts through local communication is presented. We discuss, through experimental and simulation results, two cases of floor cleaning task, and surveillance task, performed by multiple robots. Results show that the proposed scheme enables multiple service robots to perform cooperative tasks intelligently without any explicit programming.