In this paper we propose a new algorithm for robots searching a hazardous, communications-denied area to gather information using a robot fleet that has a limited number of agents. The centralized algorithm uses robot survival along search paths as a sensor event for a distributed sensor network. As agents are lost to hazards, the search behavior adjusts to prioritize agent longevity in order to maximize information gain. In the past, related work solving this problem has assumed an infinite number of agents. In contrast, we assume that the number of agents is finite. We use Bayesian inference to update target and hazard belief maps of an area using data from the probability of survival of prior agents’ paths as well as sensor readings from the agents along those paths. Using those belief maps, the algorithm can construct paths that maximize information gain, in expectation, while taking into account the predicted decrease in future information collected when losing an agent. This behavior increases the likelihood that agents survive longer, allowing them to collect more data.Using simulations with various fleet sizes and probabilities for hazards disabling agents, we compare our algorithm to work that does not account for attrition. The results show an increase in the longevity of the fleet when hazards are more effective at disabling agents. In nearly all cases, this contributes to an increased rate in information gain when the fleet size is small. Small sized fleets, in our case 10 or less agents, do not meet a threshold of collected information necessary to direct agents away from hazards. Large fleets, over 200 agents in our scenario, collect most of the information before Our algorithm causes a noticeable change in agent behavior (as compared to existing techniques). We find that the proposed method provides the greatest advantage for mid-sized fleets, between 20 and 100 agents, and when hazards have an increased probability of immobilizing agents.

Consider a sensor that reports whether or not an event has occurredsomewherealong a path, but that has no conception ofwherealong the path that event has occurred. We name this type of sensor apath-based sensorand describe the recursive Bayesian update that can be used to calculate posterior beliefs about the presence of a sensor triggering phenomenon given a path-based sensor observation. We show how the Bayesian update can be leveraged to calculate the expected Shannon information that will be gained along a particular path. We formalize two iterative information-gathering problems that result from this scenario and present path-planning algorithms to solve them. These include: 1) gathering information about the path-based sensor triggering phenomena and 2) assuming the path-based sensor triggering event is “robot destruction,” simultaneously gather information about: 1) hazards using a path-based sensor and 2) information about another environmental phenomenon using a standard sensor, such as the locations of search and rescue targets with a camera. We evaluate our methods using Monte Carlo simulations and observe that they outperform other techniques with respect to the new problems that we consider.Note to Practitioners—This work is motivated by the problem of searching for robot-destroying hazards that are otherwise invisible to the robots. That is, we can observe whether or not a robot survives a path, but, if a robot is destroyed, then we have no idea where, along the path, its destruction has occurred. A mathematically equivalent problem happens in any scenario, in which an agent is equipped with an event sensor that can only be set/triggered once, but that requires postprocessing to figure out if the sensor has been triggered or not. For example, postprocessing is needed if the determination of whether or not a biological specimen was obtained requires a manual laboratory inspection. We also consider an extension of the hazard detection problem, in which we simultaneously collect information about search-and-rescue victims using a “victim sensor,” such as a camera. In this problem, hazards indirectly affect information gathered about victims because new information about victims is lost whenever a robot is destroyed. We provide algorithms to solve these types of problems. The algorithms work even in cases with noise such that false positives and false negatives are possible. This work is useful in any application where observations take the form of a cumulative “yes” or “no” along a path.

In this paper we propose an algorithm to actively control the distance of a motor-propeller system (MPS) to a large obstacle using data from a single microphone. The method is based upon a broadband constructive/destructive interference pattern across the audible frequency band that is present when the MPS is near an obstacle. By taking the difference between the power spectrum in the obstacle-free case and the spectrum when recording near an obstacle, a broadband oscillation with respect to frequency is revealed. The frequency of this oscillation is linearly-related to the distance from the microphone to the wall. We present both static and dynamic experiments showcasing the ability of the proposed method to estimate the distance to a wall as well as actively control it.

This work describes our approach to controlling lighter-than-air agents using multimodal control via a wearable device. Tactile and gesture interfaces on a smart watch are used to control the motion and altitude of these semi-autonomous agents. The tactile interface consists of the touch screen and rotatable bezel. The gesture interface detects when the user puts his/her hand in the stop position. The touch interface controls the direction of the agents, the rotatable bezel controls the altitude set-point, and the gesture interface stops the agents. Our interactive demonstration will allow users to control a lighter-than-air (LTA) system via the multimodal wearable interface as described above.

Many types of bats use echolocation to sense their environment. Despite often have little or no visual acuity, they are able to acquire very detailed views of their surroundings through emission, receipt, and analysis of acoustic pulses. In this study autonomous navigation was examined with respect to classification of nearby terrain. The goal of this effort was to demonstrate that a bat-inspired acoustic sensor could be built, and when trained using advanced signal filtering and machine learning techniques, could be used to accurately classify terrain types for a small mobile robot. A dual channel in-air sonar was constructed using two common piezoelectric transmitter elements with 25 kHz and 32 kHz nominal center frequencies, and echo data was collected from grass, concrete, sand, and gravel terrain substrates. Higher dimension time, frequency, and time-frequency PCA scores were used to discriminate between terrain substrates. These features were used to train a support vector machine (SVM) to classify the terrain types. The SVM-based classifier was able to classify terrain types at a greater than 97% success rate using the constructed bat-inspired echolocation sensor.

In this work, we present a Stochastic Gradient Ascent (SGA) algorithm for multi-vehicle information gathering that accounts for limitations on a vehicle's control authority caused by external forces. By representing vehicle paths using a novel action space representation, rather than a state space representation, we remove the need to perform feasibility calculations on the vehicle's path. Our algorithm uses a stochastic optimization scheme by sampling perturbed action sequences around the current best known sequence to estimate the gradient of a state space information function with respect to the action sequence. Additionally, we use sequential greedy allocation to plan for multiple vehicles. Results are shown using a Navy Coastal Ocean Model (NCOM) for the Gulf of Mexico (GoM). SGA shows improvement in the amount of information gained over a greedy baseline. Additionally, we compare to Monte Carlo Tree Search (MCTS) Method, which is able to gather competitive amounts of information but is more computationally intensive than our approach.

Traditionally, localization systems for formation-keeping have required external references to a world frame, such as GPS or motion capture. Humans wishing to maintain formations with each other use visual line-of-sight, which is not possible in cluttered environments that may block their view. Our Local Positioning System (LPS) is based on ultra-wideband ranging devices and it enables teams of robots and humans to know their relative locations without line-of-sight. This demonstration shows live (real-time) local positioning between three agents (human and robot). The emphasis of this demonstration is showing the wearable interactive display that allows the human to see the local frame formation of the human and the robots.

Motion planning for multi-target autonomous search requires efficiently gathering as much information over an area as possible with an imperfect sensor. In disaster scenarios and contested environments the spatial connectivity may unexpectedly change (due to aftershock, avalanche, flood, building collapse, adversary movements, etc.) and the flight envelope may evolve as a known function of time to ensure rescue worker safety or to facilitate other mission goals. Algorithms designed to handle both expected and unexpected changes must: (1) reason over a sufficiently long time horizon to respect expected changes, and (2) replan quickly in response to unexpected changes. These ambitions are hindered by the submodularity property of mutual information, which makes optimal solutions NP-hard to compute. We present an algorithm for autonomous search in changing environments that uses a variety of techniques to improve both the speed and time horizon, including using e-admissible heuristics to speed up the search.

Quadruped locomotion offers significant advantages over wheeled locomotion for small mobile robots operating in challenging terrain. Central pattern generators (CPGs), as found in the neural circuitry of many animals, may be used to generate joint trajectories for quadruped robots. However, basic CPG-based trajectories do not explicitly consider ground contact constraints, a particular concern during turning maneuvers when foot slip is most likely to occur. An alternative approach proposed here is to use task-based CPGs such that ground contact constraints are enforced and foot velocities are explicitly controlled, resulting in stable omnidirectional locomotion. Further, incorporating Virtual Model Control with the task-based CPG trajectories improves the stability of the quadruped in hardware experiments.

The papers in this special section focus on the technology and applications supported by networked cooperative autonomous systems.

This paper proposes a reactive motion-planning approach for persistent surveillance of risk-sensitive areas by a team of unmanned aerial vehicles (UAVs). The planner, termed PARCov (Planner for Autonomous Risk-sensitive Coverage), seeks to: i) maximize the area covered by sensors mounted on each UAV; ii) provide persistent surveillance; iii) maintain high sensor data quality; and iv) reduce detection risk. To achieve the stated objectives, PARCov combines into a cost function the detection risk with an uncertainty measure designed to keep track of the regions that have been surveyed and the times they were last surveyed. PARCov reduces the uncertainty and detection risk by moving each quadcopter toward a low-cost region in its vicinity. By reducing the uncertainty, PARCov is able to increase the coverage and provide persistent surveillance. Moreover, a nonlinear optimization formulation is used to determine the optimal altitude for flying each quadcopter in order to maximize the sensor data quality while minimizing risk.

Knowing the terrain is vital for small autonomous robots traversing unstructured outdoor environments. We present a technique using 3D laser point clouds combined with RGB camera images to classify terrain into four pre-defined classes: grass, sand, concrete, and metal. Our technique first segments the point cloud into distinct regions and then applies a simple classifier to determine the classification of each region. We demonstrate three classification and four segmentation algorithms on five outdoor environments. Classification and segmentation algorithms which use more information outperform information poor combinations.

Unmanned vehicles are emerging as an attractive tool for persistent monitoring tasks of a given area, but need automated planning capabilities for effective unattended deployment. Such an automated planner needs to generate collision-free coverage paths by steering waypoints to locations that both minimize the path length and maximize the amount of information gathered along the path. The approach presented in this paper significantly extends prior work and handles motion uncertainty of an unmanned vehicle and the presence of obstacles by using a Markov Decision Process based approach to generate collision-free paths. Simulation results show that the proposed approach is robust to significant motion uncertainties and reduces the probability of collision with obstacles in the environment.

Domestic grazing animals follow simple, scalable rules to assign themselves trajectories to cover a pasture. We explain how to adapt these rules for an information gathering system based on a realistic robot motion model and Kalman-filter based evidence grid that accounts for both bandwidth and sensor limitations. Our results show that this algorithm can meet or exceed the performance of state of the art field robotics systems, particularly when scalability and robustness to failure are required.

This paper proposes a path-planning approach to enable a team of unmanned aerial vehicles (UAVs) to efficiently conduct surveillance of sensitive areas. The proposed approach, termed PARCov (Planner for Autonomous Risk-sensitive Coverage), seeks to maximize the area covered by the sensors mounted on each UAV while maintaining high sensor data quality and minimizing detection risk. PARCov leverages from swarm intelligence the idea of using simple interactions among UAVs to promote an emergent behavior that achieves the desired objectives. PARCov uses a dynamic grid to keep track of the parts of the space that have been surveyed and the times that they were last surveyed. This information is then used to move the UAVs toward areas that have not been covered in a long time. Moreover, a nonlinear optimization formulation is used to determine the altitude at which each UAV flies. The efficiency and scalability of PARCov is demonstrated in simulation using complex environments and an increasing number of UAVs to conduct risk-sensitive surveillance.

Seeding the population of an evolutionary algorithm with solutions from previous runs has proved to be useful when learning control strategies for agents operating in a complex, changing environment. It has generally been assumed that initializing a learning algorithm with previously learned solutions will be helpful if the new problem is similar to the old. We will show that this assumption sometimes does not hold for many reasonable similarity metrics. Using a more traditional machine learning perspective, we explain why seeding is sometimes not helpful by looking at the learning-experience bias produced by the previously evolved solutions.

Cooperative coevolutionary architectures provide a framework for solving complex problems by decomposing them into constituent subproblems, solving the subproblems, and then reintegrating the solutions. This paper presents a blended cooperative coevolution model which offers advantages over traditional evolutionary algorithms and currently-used cooperative coevolutionary architectures.