Frequency response method for terrain classification in autonomous ground vehicles

Sudden stop detection and automatic seating support with neural stimulation during manual wheelchair propulsion

Objective: Wheelchair safety is of great importance since falls from wheelchairs are prevalent and often have devastating consequences. We developed an automatic system to detect destabilizing events during wheelchair propulsion under real-world conditions and trigger neural stimulation to stiffen the trunk to maintain seated postures of users with paralysis.Design: Cross-over interventionSetting: Laboratory and community settingsParticipants: Three able-bodied subjects and three individuals with SCI with previously implanted neurostimulation systemsInterventions: An algorithm to detect wheelchair sudden stops was developed. This was used to randomly trigger trunk extensor stimulation during sudden stops eventsOutcome Measures: Algorithm success and false positive rates were determined. SCI users rated each condition on a seven-point Usability Rating Scale to indicate safety.Results: The system detected sudden stops with a success rate of over 93% in community settings. When used to trigger trunk neurostimulation to ensure stability, the implant recipients consistently reported feeling safer (P<.05 for 2/3 subjects) with the system while encountering sudden stops as indicated by a 1-3 point change in safety rating.Conclusion: These preliminary results suggest that this system could monitor wheelchair activity and only apply stabilizing neurostimulation when appropriate to maintain posture. Larger scale, unsupervised and longer-term trials at home and in the community are indicated. This system could be generalized and applied to individuals without an implanted stimulation by utilizing surface stimulation, or by actuating a mechanical restraint when necessary, thus allowing unrestricted trunk movements and only restraining the user when necessary to ensure safety.Trial Registration: NCT01474148.

AI-Enabled Predictive Maintenance Framework for Autonomous Mobile Cleaning Robots

Vibration is an indicator of performance degradation or operational safety issues of mobile cleaning robots. Therefore, predicting the source of vibration at an early stage will help to avoid functional losses and hazardous operational environments. This work presents an artificial intelligence (AI)-enabled predictive maintenance framework for mobile cleaning robots to identify performance degradation and operational safety issues through vibration signals. A four-layer 1D CNN framework was developed and trained with a vibration signals dataset generated from the in-house developed autonomous steam mopping robot 'Snail' with different health conditions and hazardous operational environments. The vibration signals were collected using an IMU sensor and categorized into five classes: normal operational vibration, hazardous terrain induced vibration, collision-induced vibration, loose assembly induced vibration, and structure imbalanced vibration signals. The performance of the trained predictive maintenance framework was evaluated with various real-time field trials with statistical measurement metrics. The experiment results indicate that our proposed predictive maintenance framework has accurately predicted the performance degradation and operational safety issues by analyzing the vibration signal patterns raised from the cleaning robot on different test scenarios. Finally, a predictive maintenance map was generated by fusing the vibration signal class on the cartographer SLAM algorithm-generated 2D environment map.

Recent developments in terrain identification, classification, parameter estimation for the navigation of autonomous robots

The work presents a review on ongoing researches in terrain-related challenges influencing the navigation of Autonomous Robots, specifically Unmanned Ground ones. The paper aims to highlight the recent developments in robot design and advanced computing techniques in terrain identification, classification, parameter estimation, and developing modern control strategies. The objective of our research is to familiarize the gaps and opportunities of the aforementioned areas to the researchers who are passionate to take up research in the field of autonomous robots. The paper brings recent works related to terrain strategies under a single platform focusing on the advancements in planetary rovers, rescue robots, military robots, agricultural robots, etc. Finally, this paper provides a comprehensive analysis of the related works which can bridge the AI techniques and advanced control strategies to improve navigation. The study focuses on various Deep Learning techniques and Fuzzy Logic Systems in detail. The work can be extended to develop new control schemes to improve multiple terrain navigation performance.

Low-Cost Road-Surface Classification System Based on Self-Organizing Maps

Expanding the performance and autonomous-decision capability of driver-assistance systems is critical in today’s automotive engineering industry to help drivers and reduce accident incidence. It is essential to provide vehicles with the necessary perception systems, but without creating a prohibitively expensive product. In this area, the continuous and precise estimation of a road surface on which a vehicle moves is vital for many systems. This paper proposes a low-cost approach to solve this issue. The developed algorithm resorts to analysis of vibrations generated by the tyre-rolling movement to classify road surfaces, which allows for optimizing vehicular-safety-system performance. The signal is analyzed by means of machine-learning techniques, and the classification and estimation of the surface are carried out with the use of a self-organizing-map (SOM) algorithm. Real recordings of the vibration produced by tyre rolling on six different types of surface were used to generate the model. The efficiency of the proposed model (88.54%) and its speed of execution were compared with those of other classifiers in order to evaluate its performance.

Terrain Estimation for Planetary Exploration Robots

A planetary exploration rover’s ability to detect the type of supporting surface is critical to the successful accomplishment of the planned task, especially for long-range and long-duration missions. This paper presents a general approach to endow a robot with the ability to sense the terrain being traversed. It relies on the estimation of motion states and physical variables pertaining to the interaction of the vehicle with the environment. First, a comprehensive proprioceptive feature set is investigated to evaluate the informative content and the ability to gather terrain properties. Then, a terrain classifier is developed grounded on Support Vector Machine (SVM) and that uses an optimal proprioceptive feature set. Following this rationale, episodes of high slippage can be also treated as a particular terrain type and detected via a dedicated classifier. The proposed approach is tested and demonstrated in the field using SherpaTT rover, property of DFKI (German Research Center for Artificial Intelligence), that uses an active suspension system to adapt to terrain unevenness.

What Lies Beneath One’s Feet? Terrain Classification Using Inertial Data of Human Walk

The objective of this study was to investigate if the inertial data collected from normal human walk can be used to reveal the underlying terrain types. For this purpose, we recorded the gait patterns of normal human walk on six different terrain types with variation in hardness and friction using body mounted inertial sensors. We collected accelerations and angular velocities of 40 healthy subjects with two smartphones embedded inertial measurement units (MPU-6500) attached at two different body locations (chest and lower back). The recorded data were segmented with stride based segmentation approach and 194 tempo-spectral features were computed for each stride. We trained two machine learning classifiers, namely random forest and support vector machine, and cross validated the results with 10-fold cross-validation strategy. The classification tasks were performed on indoor–outdoor terrains, hard–soft terrains, and a combination of binary, ternary, quaternary, quinary and senary terrains. From the experimental results, the classification accuracies of 97% and 92% were achieved for indoor–outdoor and hard–soft terrains, respectively. The classification results for binary, ternary, quaternary, quinary and senary class classification were 96%, 94%, 92%, 90%, and 89%, respectively. These results demonstrate that the stride data collected with the low-level signals of a single IMU can be used to train classifiers and predict terrain types with high accuracy. Moreover, the problem at hand can be solved invariant of sensor type and sensor location.

Deep Multi-Layer Perception Based Terrain Classification for Planetary Exploration Rovers

Accurate classification and identification of the detected terrain is the basis for the long-distance patrol mission of the planetary rover. But terrain measurement based on vision and radar is subject to conditions such as light changes and dust storms. In this paper, under the premise of not increasing the sensor load of the existing rover, a terrain classification and recognition method based on vibration is proposed. Firstly, the time-frequency domain transformation of vibration information is realized by fast Fourier transform (FFT), and the characteristic representation of vibration information is given. Secondly, a deep neural network based on multi-layer perception is designed to realize classification of different terrains. Finally, combined with the Jackal unmanned vehicle platform, the XQ unmanned vehicle platform, and the vibration sensor, the terrain classification comparison test based on five different terrains was completed. The results show that the proposed algorithm has higher classification accuracy, and different platforms and running speeds have certain influence on the terrain classification at the same time, which provides support for subsequent practical applications.

Terrain Awareness Using a Tracked Skid-Steering Vehicle With Passive Independent Suspensions

This paper presents a novel approach for terrain characterization based on a tracked skid-steer vehicle with a passive independent suspensions system. A set of physics-based parameters is used to characterize the terrain properties: drive motor electrical currents, the equivalent track, the power spectral density for the vertical accelerations and motor currents. Based on this feature set, the system predicts the type of terrain that the robot traverses. A wide set of experimental results acquired on various surfaces are provided to verify the study in the field, proving its effectiveness for application in autonomous robots.

Tactile surface classification for limbed robots using a pressure sensitive robot skin

This paper describes an approach to terrain identification based on pressure images generated through direct surface contact using a robot skin constructed around a high-resolution pressure sensing array. Terrain signatures for classification are formulated from the magnitude frequency responses of the pressure images. The initial experimental results for statically obtained images show that the approach yields classification accuracies [Formula: see text]. The methodology is extended to accommodate the dynamic pressure images anticipated when a robot is walking or running. Experiments with a one-legged hopping robot yield similar identification accuracies [Formula: see text]. In addition, the accuracies are independent with respect to changing robot dynamics (i.e., when using different leg gaits). The paper further shows that the high-resolution capabilities of the sensor enables similarly textured surfaces to be distinguished. A correcting filter is developed to accommodate for failures or faults that inevitably occur within the sensing array with continued use. Experimental results show using the correcting filter can extend the effective operational lifespan of a high-resolution sensing array over 6x in the presence of sensor damage. The results presented suggest this methodology can be extended to autonomous field robots, providing a robot with crucial information about the environment that can be used to aid stable and efficient mobility over rough and varying terrains.

A Comparison of RBF Neural Network Training Algorithms for Inertial Sensor Based Terrain Classification

This paper introduces a comparison of training algorithms of radial basis function (RBF) neural networks for classification purposes. RBF networks provide effective solutions in many science and engineering fields. They are especially popular in the pattern classification and signal processing areas. Several algorithms have been proposed for training RBF networks. The Artificial Bee Colony (ABC) algorithm is a new, very simple and robust population based optimization algorithm that is inspired by the intelligent behavior of honey bee swarms. The training performance of the ABC algorithm is compared with the Genetic algorithm, Kalman filtering algorithm and gradient descent algorithm. In the experiments, not only well known classification problems from the UCI repository such as the Iris, Wine and Glass datasets have been used, but also an experimental setup is designed and inertial sensor based terrain classification for autonomous ground vehicles was also achieved. Experimental results show that the use of the ABC algorithm results in better learning than those of others.