Navinda Kottege

Benjamin Tam, Fletcher Talbot, Ryan Steindl et al.

Multilegged robots have the ability to perform stable locomotion on relatively rough terrain. However, the complexity of legged robots over wheeled or tracked robots make them difficult to control. This paper presents OpenSHC (Open-source Syropod High-level Controller), a versatile high-level controller {capable of generating gaits and poses for quasi-static multilegged robots, both simulated and with real hardware implementations. With full Robot Operating System (ROS) integration, the controller can be quickly deployed on robots with different actuators and sensor payloads}. The flexibility of OpenSHC is demonstrated on the 30 degrees of freedom hexapod Bullet, analysing the energetic performance of various leg configurations, kinematic arrangements and gaits over different locomotion speeds. With OpenSHC being easily configured to different physical and locomotion specifications, a hardware-based parameter space search for optimal locomotion parameters is conducted. The experimental evaluation shows that the mammalian configuration offers lower power consumption across a range of step frequencies; with the insectoid configuration providing performance advantages at higher body velocities and increased stability at low step frequencies. OpenSHC is open-source and able to be configured for various number of joints and legs.

Nicolas Hudson, Fletcher Talbot, Mark Cox et al.

Heterogeneous teams of robots, leveraging a balance between autonomy and human interaction, bring powerful capabilities to the problem of exploring dangerous, unstructured subterranean environments. Here we describe the solution developed by Team CSIRO Data61, consisting of CSIRO, Emesent and Georgia Tech, during the DARPA Subterranean Challenge. These presented systems were fielded in the Tunnel Circuit in August 2019, the Urban Circuit in February 2020, and in our own Cave event, conducted in September 2020. A unique capability of the fielded team is the homogeneous sensing of the platforms utilised, which is leveraged to obtain a decentralised multi-agent SLAM solution on each platform (both ground agents and UAVs) using peer-to-peer communications. This enabled a shift in focus from constructing a pervasive communications network to relying on multi-agent autonomy, motivated by experiences in early circuit events. These experiences also showed the surprising capability of rugged tracked platforms for challenging terrain, which in turn led to the heterogeneous team structure based on a BIA5 OzBot Titan ground robot and an Emesent Hovermap UAV, supplemented by smaller tracked or legged ground robots. The ground agents use a common CatPack perception module, which allowed reuse of the perception and autonomy stack across all ground agents with minimal adaptation.

Ahmadreza Ahmadi, Tønnes Nygaard, Navinda Kottege et al.

Legged robots are popular candidates for missions in challenging terrains due to the wide variety of locomotion strategies they can employ. Terrain classification is a key enabling technology for autonomous legged robots, as it allows the robot to harness their innate flexibility to adapt their behaviour to the demands of their operating environment. In this paper, we show how highly capable machine learning techniques, namely gated recurrent neural networks, allow our target legged robot to correctly classify the terrain it traverses in both supervised and semi-supervised fashions. Tests on a benchmark data set shows that our time-domain classifiers are well capable of dealing with raw and variable-length data with small amount of labels and perform to a level far exceeding the frequency-domain classifiers. The classification results on our own extended data set opens up a range of high-performance behaviours that are specific to those environments. Furthermore, we show how raw unlabelled data is used to improve significantly the classification results in a semi-supervised model.

Bethany Lu, Benjamin Tam, Navinda Kottege

Legged robots traversing in confined environments could find their only path is blocked by obstacles. In circumstances where the obstacles are movable, a multilegged robot can manipulate the obstacles using its legs to allow it to continue on its path. We present a method for a hexapod robot to autonomously generate manipulation trajectories for detected obstacles. Using a RGB-D sensor as input, the obstacle is extracted from the environment and filtered to provide key contact points for the manipulation algorithm to calculate a trajectory to move the obstacle out of the path. Experiments on a 30 degree of freedom hexapod robot show the effectiveness of the algorithm in manipulating a range of obstacles in a 3D environment using its front legs.

Bhanuka Silva, Navinda Kottege

Within the scope of the paper, electromechanical and thermodynamic models are derived for a series elastic actuator and open loop and closed loop torque bandwidth parameters are analysed considering the thermodynamic behaviour of the actuator. It was observed that the closed loop torque bandwidth of the electromechanical subsystem of the actuator was not accessible in the entire torque reference amplitude range due to thermodynamic limitations. Therefore, a stator winding temperature estimation based adaptive controller is utilised and analysed to improve the accessibility of the controller based torque bandwidth. This paper implements the methodology on a HEBI Robotics X5-9 actuator as a case study.

Ryan Steindl, Thomas Molnar, Fletcher Talbot et al.

This paper introduces Bruce, the CSIRO Dynamic Hexapod Robot capable of autonomous, dynamic locomotion over difficult terrain. This robot is built around Apptronik linear series elastic actuators, and went from design to deployment in under a year by using approximately 80\% 3D printed structural (joints and link) parts. The robot has so far demonstrated rough terrain traversal over grass, rocks and rubble at 0.3m/s, and flat-ground speeds up to 0.5m/s. This was achieved with a simple controller, inspired by RHex, with a central pattern generator, task-frame impedance control for individual legs and no foot contact detection. The robot is designed to move at up to 1.0m/s on flat ground with appropriate control, and was deployed into the the DARPA SubT Challenge Tunnel circuit event in August 2019.

Thomas Hines, Kazys Stepanas, Fletcher Talbot et al.

This paper presents an autonomous navigation system for ground robots traversing aggressive unstructured terrain through a cohesive arrangement of mapping, deliberative planning and reactive behaviour modules. All systems are aware of terrain slope, visibility and vehicle orientation, enabling robots to recognize, plan and react around unobserved areas and overcome negative obstacles, slopes, steps, overhangs and narrow passageways. This is one of pioneer works to explicitly and simultaneously couple mapping, planning and reactive components in dealing with negative obstacles. The system was deployed on three heterogeneous ground robots for the DARPA Subterranean Challenge, and we present results in Urban and Cave environments, along with simulated scenarios, that demonstrate this approach.

Russell Buchanan, Tirthankar Bandyopadhyay, Marko Bjelonic et al.

Legged robots have the ability to adapt their walking posture to navigate confined spaces due to their high degrees of freedom. However, this has not been exploited in most common multilegged platforms. This paper presents a deformable bounding box abstraction of the robot model, with accompanying mapping and planning strategies, that enable a legged robot to autonomously change its body shape to navigate confined spaces. The mapping is achieved using robot-centric multi-elevation maps generated with distance sensors carried by the robot. The path planning is based on the trajectory optimisation algorithm CHOMP which creates smooth trajectories while avoiding obstacles. The proposed method has been tested in simulation and implemented on the hexapod robot Weaver, which is 33cm tall and 82cm wide when walking normally. We demonstrate navigating under 25cm overhanging obstacles, through 70cm wide gaps and over 22cm high obstacles in both artificial testing spaces and realistic environments, including a subterranean mining tunnel.

Michael Thoreau, Navinda Kottege

Tracking by detection is a common approach to solving the Multiple Object Tracking problem. In this paper we show how learning a deep similarity metric can improve three key aspects of pedestrian tracking on a multiple object tracking benchmark. We train a convolutional neural network to learn an embedding function in a Siamese configuration on a large person re-identification dataset. The offline-trained embedding network is integrated in to the tracking formulation to improve performance while retaining real-time performance. The proposed tracker stores appearance metrics while detections are strong, using this appearance information to: prevent ID switches, associate tracklets through occlusion, and propose new detections where detector confidence is low. This method achieves competitive results in evaluation, especially among online, real-time approaches. We present an ablative study showing the impact of each of the three uses of our deep appearance metric.

Raja Jurdak, Alberto Elfes, Branislav Kusy et al.

The global movement of people and goods has increased the risk of biosecurity threats and their potential to incur large economic, social, and environmental costs. Conventional manual biosecurity surveillance methods are limited by their scalability in space and time. This article focuses on autonomous surveillance systems, comprising sensor networks, robots, and intelligent algorithms, and their applicability to biosecurity threats. We discuss the spatial and temporal attributes of autonomous surveillance technologies and map them to three broad categories of biosecurity threat: (i) vector-borne diseases; (ii) plant pests; and (iii) aquatic pests. Our discussion reveals a broad range of opportunities to serve biosecurity needs through autonomous surveillance.